A model is still running

A model is still running or downloading. Leaving will stop the current operation. Stay on this page to let it finish.

Skip to content
LocalAI.toolsBeta

Join the waitlist

Get an email when the new apps and model runs launch.

No account needed.

Qwen3 ForcedAligner Desktop

Align an existing transcript to mono audio and estimate word/character start and end times using the official non-autoregressive 5000-class timestamp head. Does not transcribe unknown speech.

Desktop inferencePrivate · Runs on your computer
About this model

Align an existing transcript to mono audio and estimate word/character start and end times using the official non-autoregressive 5000-class timestamp head. Does not transcribe unknown speech.

License
Apache-2.0
Model released
Jan 29, 2026
Source repository created
Jan 28, 2026
Model weights · FP32
1.8 GB

Best for

  • Correct subtitle word timing against an existing transcript
  • Inspect local transcript/audio alignment with human review

Keep in mind

  • Requires an accurate supplied transcript; wrong or missing words can receive plausible but incorrect timestamps. Not transcription or speech recognition.
  • Adapter accepts up to30seconds of normalized mono16kHz audio and4000Unicode characters. These are conservative application limits, not the upstream five-minute maximum.
  • Upstream accepts11languages, but local quality evidence is an English pilot from one speech recording with clean/noise/shift conditions. No broad multilingual accuracy claim.
  • Timestamp quantization is80ms. Zero-duration spans, overlaps, incorrect word splitting and boundary errors remain possible; review subtitle timing.
  • Reference quality metric measures agreement with saved ElevenLabs Scribe v2 boundaries, not human ground truth.
  • CPU audio features/tokenization/timestamp decoding remain outside Metal; GPU acceleration is partial. Only this AppleSiliconMac runtime tested.
  • No incremental alignment output is exposed by the pinned nonautoregressive forward/decode API; generateStream is unavailable.
  • No reliable speech-presence or transcript-truth guarantee: silence or mismatched transcripts may produce fabricated timing. Not suitable for identity, forensic or high-impact decisions.

Processed on your deviceYour input is processed locally in the desktop app. Model and runtime files download from external hosts.

Target platforms · verification required

  • macos · aarch64, x86_64
  • windows · x86_64

Recommended configuration

FP32 · mps (verified platform only)

Sources & technical details
Original model
https://huggingface.co/Qwen/Qwen3-ForcedAligner-0.6B
Release verification
Verified
Release evidence
Official January29 release announcement explicitly names ForcedAligner0.6B. Official HF-native adaptation released June26 is an artifact adaptation, not a new original checkpoint release.
Source updated
Jan 30, 2026
Native model files
Qwen/Qwen3-ForcedAligner-0.6B-hf
Files updated
Jun 26, 2026
Artifact revision
c07281df297b9905d24a508279258cccf987a064

Desktop app

Sample result

Qwen3 ForcedAligner 0.6B Desktop

Run Qwen3 ForcedAligner Desktop on your computer

Download the desktop app to load this model, provide input and view results locally.

Public desktop downloads are coming soon.

Device requirements

macOS Apple SiliconCPU / FP32Tested16.0 GiBRAM estimate
macOS Apple SiliconMPS / FP32Tested16.0 GiBRAM estimate
Windows Intel / AMDCPU / FP32Unverified16.0 GiBRAM estimate
macOS Intel / AMDCPU / FP32Unverified16.0 GiBRAM estimate
If your device falls short

MemoryAdvisory only: close other apps, reduce image/tile/token/batch settings or choose a smaller verified model/precision. CPU can need more RAM than GPU.

GPUIf the selected GPU runtime fails, existing CPU fallback may recover when enabled, but CPU has its own memory/runtime needs.

PlatformUse a compatible OS/architecture/runtime or another model. These estimates do not add hardware detection or block runs.

Benchmark evidence

Saved desktop measurements apply to the recorded hardware and runtime. Fallback cases report the device that actually ran; they do not prove GPU support on the requested device.

Qwen3 ForcedAligner 0.6B Desktop

54 configurations tested
Passed54
Failed0
TestedNov 3, 2026
Runtimetransformers 5.22.1

Quality criteria

{ "scope": "Initial English forced-alignment screening, frozen before inference; no broad multilingual claim", "gates": { "exactTranscriptTokenCoverage": true, "allTimesFinite": true, "nonnegativeMonotonicSpans": true, "endWithinAudioToleranceSeconds": 0.24, "maximumZeroDurationFraction": 0.1, "maximumReferenceBoundaryMAESeconds": 0.2, "maximumRealTimeFactor": 1 }, "negativeCases": [ "empty transcript rejected", "nonfinite audio rejected", "unsupported language rejected" ], "performancePolicy": "At least realtime warm alignment on a short clean English recording. Reference boundary quality requires an independent saved reference before support can be advertised." }

PrecisionDeviceResultPlatformInference / loadTask metric
FP32mpsAcceleratedPassedVery fastmacos · aarch64131 ms / 0 ms0.011 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64133 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64134 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64132 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64132 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64132 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64133 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64134 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64134 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64135 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64136 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64136 ms / 0 ms0.012 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64209 ms / 0 ms0.018 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64216 ms / 0 ms0.019 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64218 ms / 0 ms0.019 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64301 ms / 849 ms0.027 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64320 ms / 847 ms0.029 RTF
FP32mpsAcceleratedPassedVery fastmacos · aarch64351 ms / 848 ms0.032 RTF
FP32cpuPassedVery fastmacos · aarch64342 ms / 0 ms0.030 RTF
FP32cpuPassedVery fastmacos · aarch64347 ms / 0 ms0.030 RTF
FP32cpuPassedVery fastmacos · aarch64333 ms / 0 ms0.030 RTF
FP32cpuPassedVery fastmacos · aarch64336 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64338 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64341 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64356 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64341 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64360 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64361 ms / 0 ms0.031 RTF
FP32cpuPassedVery fastmacos · aarch64349 ms / 0 ms0.032 RTF
FP32cpuPassedVery fastmacos · aarch64351 ms / 0 ms0.032 RTF
FP32cpuPassedVery fastmacos · aarch64370 ms / 0 ms0.032 RTF
FP32cpuPassedVery fastmacos · aarch64367 ms / 0 ms0.033 RTF
FP32cpuPassedVery fastmacos · aarch64369 ms / 0 ms0.034 RTF
FP32cpuPassedVery fastmacos · aarch64373 ms / 474 ms0.034 RTF
FP32cpuPassedVery fastmacos · aarch64413 ms / 0 ms0.036 RTF
FP32cpuPassedVery fastmacos · aarch64402 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64405 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64406 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64429 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64410 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64411 ms / 0 ms0.037 RTF
FP32cpuPassedVery fastmacos · aarch64416 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64417 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64436 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64436 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64438 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64419 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64421 ms / 0 ms0.038 RTF
FP32cpuPassedVery fastmacos · aarch64451 ms / 0 ms0.039 RTF
FP32cpuPassedVery fastmacos · aarch64440 ms / 1330 ms0.040 RTF
FP32cpuPassedVery fastmacos · aarch64448 ms / 515 ms0.041 RTF
FP32cpuPassedVery fastmacos · aarch64450 ms / 491 ms0.041 RTF
FP32cpuPassedVery fastmacos · aarch64451 ms / 1349 ms0.041 RTF
FP32cpuPassedVery fastmacos · aarch64463 ms / 487 ms0.042 RTF
Test setup & sample details
Quality gate
{ "scope": "Initial English forced-alignment screening, frozen before inference; no broad multilingual claim", "gates": { "exactTranscriptTokenCoverage": true, "allTimesFinite": true, "nonnegativeMonotonicSpans": true, "endWithinAudioToleranceSeconds": 0.24, "maximumZeroDurationFraction": 0.1, "maximumReferenceBoundaryMAESeconds": 0.2, "maximumRealTimeFactor": 1 }, "negativeCases": [ "empty transcript rejected", "nonfinite audio rejected", "unsupported language rejected" ], "performancePolicy": "At least realtime warm alignment on a short clean English recording. Reference boundary quality requires an independent saved reference before support can be advertised." }
Tested
2026-11-03T17:28:04.763746+00:00
Environment
{ "hardware": "Apple M3 Max", "os": "27.2", "architecture": "aarch64", "runtime": "Transformers5.22.1/Torch2.10.0", "interpreterScope": "Existing app-owned isolated CPython3.13.18 screening environment; separate managed packaged test required" }
Runtime
transformers 5.22.1
Artifact revision
c07281df297b9905d24a508279258cccf987a064
Fixture
{
  "file": "model-samples/desktop/qwen3-forced-aligner/clean.wav",
  "url": "https://assets.localai.tools/samples/v1/files/ba1a561773d6b6e55c7b9d326d9aa792f6d3a6f430556b83a78f78a2a4cb25e1/clean.wav",
  "size": 352044,
  "sha1": "ce87ae892606eecb8ef65803ed7c4d1ea13ca437",
  "sha256": "ba1a561773d6b6e55c7b9d326d9aa792f6d3a6f430556b83a78f78a2a4cb25e1"
}
Rerun when
Exact artifact/runtime revision; Audio/transcript/reference checksum or quality gates; Task preprocessing/decoding or backend behavior
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 515.4390420066193,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 457.42287501343526,
      "inferenceMs": 448.0288750200998,
      "realTimeFactor": 0.04072989772909998,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 972.8619170200545,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 408.8097909989301,
      "inferenceMs": 405.0346250005532,
      "realTimeFactor": 0.036821329545504836,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 408.8097909989301,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 441.9694590033032,
      "inferenceMs": 435.89537500520237,
      "realTimeFactor": 0.037903945652626295,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 441.9694590033032,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 423.17662498680875,
      "inferenceMs": 419.29445799905807,
      "realTimeFactor": 0.03811767799991437,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 423.17662498680875,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 419.62533397600055,
      "inferenceMs": 415.7910840003751,
      "realTimeFactor": 0.03779918945457956,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 419.62533397600055,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 439.4640829996206,
      "inferenceMs": 435.52358300075866,
      "realTimeFactor": 0.037871615913109447,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2164.2741250107065,
      "totalMs": 439.4640829996206,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 491.2973339960445,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 460.0634580128826,
      "inferenceMs": 450.2905420085881,
      "realTimeFactor": 0.040935503818962556,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 951.3607920089271,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 414.39066600287333,
      "inferenceMs": 410.18162498949096,
      "realTimeFactor": 0.03728923863540827,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 414.39066600287333,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 455.1565000147093,
      "inferenceMs": 450.87670799694024,
      "realTimeFactor": 0.039206670260603496,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 455.1565000147093,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 424.7031249979045,
      "inferenceMs": 420.8665000041947,
      "realTimeFactor": 0.038260590909472245,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 424.7031249979045,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 415.3417919878848,
      "inferenceMs": 411.47266601910815,
      "realTimeFactor": 0.037406606001737105,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 415.3417919878848,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 432.3914580163546,
      "inferenceMs": 428.5819169890601,
      "realTimeFactor": 0.0372679927816574,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 2040.405625011772,
      "totalMs": 432.3914580163546,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 487.13337499066256,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 487.0876669883728,
      "inferenceMs": 462.8347090038005,
      "realTimeFactor": 0.04207588263670914,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 974.2210419790354,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 420.1860410103109,
      "inferenceMs": 416.5407080145087,
      "realTimeFactor": 0.037867337092228066,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 420.1860410103109,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 441.7268749966752,
      "inferenceMs": 437.6395409926772,
      "realTimeFactor": 0.0380556122602328,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 441.7268749966752,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 410.2759170054924,
      "inferenceMs": 406.33887500735,
      "realTimeFactor": 0.03693989772794091,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 410.2759170054924,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 405.9830000041984,
      "inferenceMs": 402.3382920131553,
      "realTimeFactor": 0.0365762083648323,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 405.9830000041984,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 416.3032910146285,
      "inferenceMs": 412.5089170120191,
      "realTimeFactor": 0.03587034060974079,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1937.6122920075431,
      "totalMs": 416.3032910146285,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 1,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 473.97383401403204,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 382.057957991492,
      "inferenceMs": 372.8932499943767,
      "realTimeFactor": 0.03389938636312515,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 856.031792005524,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 370.3995000105351,
      "inferenceMs": 367.1842920011841,
      "realTimeFactor": 0.033380390181925824,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 370.3995000105351,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 372.97179200686514,
      "inferenceMs": 369.6766250068322,
      "realTimeFactor": 0.032145793478854975,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 372.97179200686514,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 351.60525000537746,
      "inferenceMs": 348.6058750131633,
      "realTimeFactor": 0.031691443183014846,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 351.60525000537746,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 343.66416701232083,
      "inferenceMs": 340.7387080078479,
      "realTimeFactor": 0.030976246182531628,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 343.66416701232083,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 363.5937499930151,
      "inferenceMs": 360.39699998218566,
      "realTimeFactor": 0.03133886956366832,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 1994.8367080069147,
      "totalMs": 363.5937499930151,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 1349.3350839999039,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 460.34900000086054,
      "inferenceMs": 451.35962500353344,
      "realTimeFactor": 0.0410326931821394,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 1809.6840840007644,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 341.27933299168944,
      "inferenceMs": 338.3848340017721,
      "realTimeFactor": 0.030762257636524737,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 341.27933299168944,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 364.5909579936415,
      "inferenceMs": 361.34150001453236,
      "realTimeFactor": 0.03142100000126368,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 364.5909579936415,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 372.7104590216186,
      "inferenceMs": 369.43000002065673,
      "realTimeFactor": 0.03358454545642334,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 372.7104590216186,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 354.39845800283365,
      "inferenceMs": 351.4739170204848,
      "realTimeFactor": 0.031952174274589525,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 354.39845800283365,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 359.38174999319017,
      "inferenceMs": 356.4150840102229,
      "realTimeFactor": 0.03099261600088895,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3706.7139169957954,
      "totalMs": 359.38174999319017,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 1329.7146660042927,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 450.0681659847032,
      "inferenceMs": 439.88554101088084,
      "realTimeFactor": 0.039989594637352806,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 1779.782831988996,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 344.187165988842,
      "inferenceMs": 341.0860000003595,
      "realTimeFactor": 0.031007818181850864,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 344.187165988842,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 350.3464170207735,
      "inferenceMs": 347.1394579974003,
      "realTimeFactor": 0.0301860398258609,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 350.3464170207735,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 339.2552080040332,
      "inferenceMs": 336.3931250059977,
      "realTimeFactor": 0.03058119318236343,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 339.2552080040332,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 336.41237500705756,
      "inferenceMs": 333.3947920182254,
      "realTimeFactor": 0.03030861745620231,
      "audioSeconds": 11,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 336.41237500705756,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "cpu/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "PyTorch CPU",
      "acceleration": "cpu",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 344.77783300098963,
      "inferenceMs": 341.5556249965448,
      "realTimeFactor": 0.02970048913013433,
      "audioSeconds": 11.5,
      "outputPlacement": "cpu",
      "sampleRate": 16000,
      "processInitMs": 3765.774917003,
      "totalMs": 344.77783300098963,
      "device": "cpu",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 848.3570420066826,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 380.836916010594,
      "inferenceMs": 351.05566601851024,
      "realTimeFactor": 0.0319141514562282,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505059328,
      "processInitMs": 2672.269792004954,
      "totalMs": 1229.1939580172766,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 140.44737501535565,
      "inferenceMs": 135.16887501464225,
      "realTimeFactor": 0.012288079546785659,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505042944,
      "processInitMs": 2672.269792004954,
      "totalMs": 140.44737501535565,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 234.4101250055246,
      "inferenceMs": 215.85033400333486,
      "realTimeFactor": 0.018769594261159553,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2672.269792004954,
      "totalMs": 234.4101250055246,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 138.8297500088811,
      "inferenceMs": 131.98262499645352,
      "realTimeFactor": 0.011998420454223047,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2672.269792004954,
      "totalMs": 138.8297500088811,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 138.91087501542643,
      "inferenceMs": 132.22641698666848,
      "realTimeFactor": 0.012020583362424408,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2672.269792004954,
      "totalMs": 138.91087501542643,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 139.85874998616055,
      "inferenceMs": 134.0589580067899,
      "realTimeFactor": 0.0116573006962426,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2672.269792004954,
      "totalMs": 139.85874998616055,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 847.3243339976761,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 352.5462500110734,
      "inferenceMs": 319.5676250034012,
      "realTimeFactor": 0.02905160227303647,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505059328,
      "processInitMs": 2363.5727500077337,
      "totalMs": 1199.8705840087496,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 143.4674579941202,
      "inferenceMs": 136.43204199615866,
      "realTimeFactor": 0.012402912908741697,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505059328,
      "processInitMs": 2363.5727500077337,
      "totalMs": 143.4674579941202,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 241.57225000089966,
      "inferenceMs": 218.01712500746362,
      "realTimeFactor": 0.018958010870214228,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2363.5727500077337,
      "totalMs": 241.57225000089966,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 138.93075002124533,
      "inferenceMs": 132.8381659986917,
      "realTimeFactor": 0.012076196908971971,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2363.5727500077337,
      "totalMs": 138.93075002124533,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 138.66799999959767,
      "inferenceMs": 133.62116599455476,
      "realTimeFactor": 0.012147378726777706,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2363.5727500077337,
      "totalMs": 138.66799999959767,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 140.10954200057313,
      "inferenceMs": 132.8815000015311,
      "realTimeFactor": 0.0115549130436114,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2363.5727500077337,
      "totalMs": 140.10954200057313,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 848.8075839995872,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 329.1185420239344,
      "inferenceMs": 300.74145799153484,
      "realTimeFactor": 0.027340132544684984,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505059328,
      "processInitMs": 2346.4168339851312,
      "totalMs": 1177.9261260235216,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 141.6101670183707,
      "inferenceMs": 135.84187501692213,
      "realTimeFactor": 0.01234926136517474,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4505059328,
      "processInitMs": 2346.4168339851312,
      "totalMs": 141.6101670183707,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 228.99274999508634,
      "inferenceMs": 208.74887498212047,
      "realTimeFactor": 0.01815207608540178,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2346.4168339851312,
      "totalMs": 228.99274999508634,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05318181818181815,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 140.12429202557541,
      "inferenceMs": 133.96245799958706,
      "realTimeFactor": 0.012178405272689733,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2346.4168339851312,
      "totalMs": 140.12429202557541,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.05045454545454546,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 138.81574998958968,
      "inferenceMs": 132.47708301059902,
      "realTimeFactor": 0.01204337118278173,
      "audioSeconds": 11,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2346.4168339851312,
      "totalMs": 138.81574998958968,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "boundaryMAESeconds": 0.055000000000000056,
      "zeroDurationFraction": 0,
      "referenceType": "ElevenLabsScribeV2Agreement",
      "passed": true
    },
    "runtime": {
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 0,
      "placement": {
        "mps:0/torch.float32": 708
      },
      "parameterDtype": "torch.float32",
      "provider": "Metal/MPS",
      "acceleration": "partial",
      "accelerationDetails": "Model parameters and forward tensors on selected device; audio features, transcript splitting and timestamp decoding on CPU.",
      "durationMs": 137.62887500342913,
      "inferenceMs": 130.5495830019936,
      "realTimeFactor": 0.01135213765234727,
      "audioSeconds": 11.5,
      "outputPlacement": "mps:0",
      "sampleRate": 16000,
      "driverAllocatedMemoryBytes": 4574068736,
      "processInitMs": 2346.4168339851312,
      "totalMs": 137.62887500342913,
      "device": "mps",
      "precision": "fp32",
      "cpuThreads": 4,
      "artifactRevision": "c07281df297b9905d24a508279258cccf987a064",
      "grade": "excellent"
    }
  }
]

Something not quite right?

Tell us what happened with Qwen3 ForcedAligner Desktop. Loading, results, or the page itself—we’re here for all of it.

Attachments Optional

Up to 3 files · 50 MB each

Browser, OS, CPU/GPU, memory, screen, storage, and network details are included automatically. Model inputs, results, and browsing history are not collected.