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.

KittenTTS Nano 0.8 INT8

Read English text aloud with a compact speech-synthesis model and preset voices. Returns 24 kHz mono audio; designed for short text inputs.

Browser inferencePowered by WebAI · Private · Unlimited runs
About this model

Read English text aloud with a compact speech-synthesis model and preset voices. Returns 24 kHz mono audio; designed for short text inputs.

Created by
KittenML
License
Apache-2.0
Model released
Unknown
Source repository created
Feb 19, 2026
Model weights · INT8
24.4 MB

Best for

  • Short English narration and voice prompts.
  • Local accessibility and speech synthesis prototypes.

Keep in mind

  • English only; names, abbreviations, numbers and punctuation can be mispronounced.
  • Small quantized model may sound unnatural or omit words; review generated speech.
  • Text is phonemized with eSpeak NG and split into chunks of at most 397 phoneme tokens; chunk boundaries can affect prosody.
  • Synthetic speech must not be used to impersonate people or mislead listeners.
  • Browser performance and backend compatibility depend on hardware; advertised pairs are benchmark-specific.

Processed on your deviceInference runs in this browser. Input/output sharing requires consent for each run. Basic run logs include device information and approximate connection location. Model and runtime files download from external hosts. Saved downloads may be cleared by your browser.

Sources & technical details
Original model
KittenML/kitten-tts-nano-0.8-int8
Release verification
Unknown
Release evidence
Exact original checkpoint public release date has not been verified. Repository creation, training, paper, artifact conversion and update dates are not release evidence.
Source updated
Feb 20, 2026
Browser model files
KittenML/kitten-tts-nano-0.8-int8
Files updated
Feb 20, 2026
Artifact revision
84781d74e29ee25217551556398b42f80593a813

Input

TEXT

Sample result

KittenTTS Nano 0.8 INT8

Settings

Selects a preset synthetic speaking voice. Each listed value selects a different voice identity. There is no higher or lower quality ordering; pronunciation and sound vary by voice and text.
Changes the generated speaking rate. Lower values speak more slowly and usually produce longer audio. Higher values speak faster and may reduce intelligibility. A value of 1 uses the normal rate.

Output

Your generated speech will appear here

Enter text and choose a reference recording, or try the sample.

Benchmark evidence

Saved worker evidence is separate from benchmarks run on your device. It does not guarantee speed or quality on every browser.

KittenTTS Nano 0.8 INT8

2 configurations tested
Passed1
Failed1
TestedOct 31, 2026
RuntimeONNX Runtime Web / phonemizer 1.24.3 / 1.2.1

Quality criteria

Finite non-silent audio, plausible duration, reference waveform envelope correlation >= 0.95 and duration ratio 0.95–1.05, ASR WER <= 0.25 with no repeated gibberish.

PrecisionDeviceResult
INT8CPU · WASMPassedFast
INT8GPU · WebGPUFailed—
Test setup & sample details
Quality gate
Finite non-silent audio, plausible duration, reference waveform envelope correlation >= 0.95 and duration ratio 0.95–1.05, ASR WER <= 0.25 with no repeated gibberish.
Tested
2026-10-31T13:44:56.369Z
Environment
Chrome 156.0.8078.79 on darwin/arm64; isolated fresh process per case; ORT 1.24.3; phonemizer 1.2.1; fflate 0.8.2
Runtime
ONNX Runtime Web / phonemizer 1.24.3 / 1.2.1
Artifact revision
84781d74e29ee25217551556398b42f80593a813
Fixture
{
  "file": "model-samples/kitten-tts-nano-0.8-int8.json",
  "url": "https://f005.backblazeb2.com/file/hubters-webai-public/assets/benchmarks/kitten-tts-nano-0.8-int8.json",
  "size": 241,
  "sha1": "0d42895ff75b33dd23aba91af415026880bb0afd",
  "reference": {
    "file": "model-samples/kitten-tts-nano-0.8-int8-reference.wav",
    "url": "https://f005.backblazeb2.com/file/hubters-webai-public/assets/benchmarks/kitten-tts-nano-0.8-int8-reference.wav",
    "size": 336080,
    "sha1": "befef1048a243a3f5a24e534989171fe2191064b",
    "samplingRate": 24000,
    "license": "CC0-1.0"
  }
}
Rerun when
artifact revision change; ONNX Runtime or phonemizer version change; fixture or quality-gate change; browser/backend behavior change
Per-case quality & failure details
[
  {
    "precision": "int8",
    "device": "wasm",
    "status": "pass",
    "quality": {
      "finite": true,
      "rms": 0.10774327000097104,
      "envelopeCorrelation": 0.9988985696795146,
      "durationRatio": 1,
      "transcript": "The quick brown fox jumps over the lazy dog.",
      "wordErrorRate": 0,
      "repetitive": false,
      "passed": true
    },
    "runtime": {
      "loadMs": 6977,
      "inferenceMs": 1407.4000000059605,
      "totalMs": 18583.59999999404,
      "grade": "good",
      "secondsPerAudioSecond": 0.40211428571598873
    }
  },
  {
    "precision": "int8",
    "device": "webgpu",
    "status": "fail",
    "quality": {
      "finite": true,
      "rms": 0.08295988897966146,
      "envelopeCorrelation": 0.8736143928330252,
      "durationRatio": 1.0071428571428571,
      "transcript": "The kitten-lounff ox jumped over the lazy dog.",
      "wordErrorRate": 0.4444444444444444,
      "repetitive": false,
      "passed": false
    },
    "runtime": {
      "loadMs": 8249.90000000596,
      "inferenceMs": 3537.7999999821186,
      "totalMs": 24599,
      "grade": "slow",
      "secondsPerAudioSecond": 1.003631205668686
    },
    "failure": "Generated audio failed the fixed native-reference/intelligibility quality gate"
  }
]

Something not quite right?

Tell us what happened with KittenTTS 0.8. 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.