omnivoice

library0.1.5pypypi✓ verified 89d ago

OmniVoice is a zero-shot text-to-speech library using diffusion language models. It supports multilingual TTS with voice cloning from short audio samples. Current version 0.1.5, actively maintained. Requires Python >= 3.10.

ai-ml llm-agents

pip install omnivoice

github ↗package ↗

INSTALL

IMPORT

SIG · OMNIVOICE

omnivoice

ai-mlpythonv0.1.5

harness data pending

Install & Compatibility

Where this runs

No compatibility data collected yet for this library.

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

OmniVoice

✓ from omnivoice import OmniVoice

✗ from omnivoice.model import OmniVoice

Direct import path changed in early versions.

infer

✓ from omnivoice import infer

✗ from omnivoice.inference import infer

Inference function available at top-level.

Basic TTS inference with voice cloning.

from omnivoice import OmniVoice, infer # Load model model = OmniVoice.from_pretrained("k2-fsa/OmniVoice") # Synthesize speech audio = infer(model, text="Hello world", reference_audio="ref.wav", reference_text="The quick brown fox") # Save to file import torchaudio torchaudio.save("output.wav", audio.unsqueeze(0), 24000)

Debug

Known issues

breakingModel loading without internet will fail if cache is missing. Use local pretrained path explicitly.

fix

Set OMNIVOICE_CACHE_DIR or download model files manually.

affects: <=0.1.5

gotchaReference audio must be monophonic and at 24kHz sample rate. Mismatch causes quality degradation.

fix

Resample audio to 24000 Hz and convert to mono before passing.

affects: all

gotchaInference on MPS (Apple Silicon) may fail due to unsupported operations. Use CPU or CUDA.

fix

Set device='cpu' explicitly when using MPS.

affects: <=0.1.5

deprecatedThe `load_asr` argument in model loading is deprecated. ASR model is now loaded automatically.

fix

Remove `load_asr=True` from `OmniVoice.from_pretrained`.

affects: >=0.1.5

Errors

Common errors & fixes

RuntimeError: Audio length mismatch

Reference audio and text lengths do not align, or audio is too long (>30s recommended).

fix

Trim reference audio to 3-30 seconds and ensure the text corresponds exactly.

AttributeError: module 'torchaudio' has no attribute 'resample'

torchaudio version is too old (<0.12) for resample function.

fix

Install torchaudio >= 0.12: pip install --upgrade torchaudio

ImportError: cannot import name 'OmniVoice' from 'omnivoice'

Incorrect import path; older documentation showed wrong path.

fix

Use 'from omnivoice import OmniVoice' instead of 'from omnivoice.model import OmniVoice'.

ValueError: The truth value of an array with more than one element is ambiguous

Passing stereo audio as reference; expects mono.

fix

Convert reference audio to mono with torchaudio.functional.to_mono().

FileNotFoundError: No such file or directory: 'path/to/model'

Model not downloaded or cache path misconfigured.

fix

Ensure internet connection for first download, or set OMNIVOICE_CACHE_DIR to a valid path.

Upgrade

Version history

0.1.5latest on PyPI · released Apr 28, 2026

Audit

Dependencies

torchrequiredCore dependency for model inference and training.

torchaudiorequiredAudio processing backend; resampling and loading audio files.

transformersrequiredRequired for tokenizer and model components.

Agent activity

21 hits · last 30 days

node

Amazon

Anthropic

OpenAI (training)

Resources

githubhttps://github.com/k2-fsa/OmniVoice ↗

packagehttps://pypi.org/project/omnivoice/ ↗