audiosronnx-voicefixer
ONNX exports of VoiceFixer (Liu et al., MIT) for
the voicefixer engine in
audiosronnx.
VoiceFixer is general speech restoration at 44.1 kHz. Where most engines target one
defect, this one is trained to undo noise, reverberation, clipping and bandwidth loss
together. Two stages, the same shape as the sidon engine: a ResUNet predicts a clean
log-mel and a TFGAN vocoder resynthesises the waveform.
analysis : mel_orig[1, 1, 501, 128] -> mel[1, 1, 501, 128]
vocoder : mel[1, 1, 501, 128] -> wav[1, 1, ~221k]
| File | Purpose | Size |
|---|---|---|
analysis.onnx | ResUNet mel predictor | 282 MB |
vocoder.onnx | TFGAN vocoder | 133 MB |
mel_filters.npy | 1025x128 htk mel filterbank for the numpy front-end | 0.5 MB |
The STFT (n_fft 2048, hop 441), the mel projection and the 10**clip(x, max=5) log inverse
all run in numpy inside audiosronnx.
Verified against torch on real speech: analysis correlation 1.00000000, the pair end-to-end 0.99999996 (71.4 dB SNR).
Notes
- Fixed window. The analysis graph is length-specialised, so it takes a fixed 5 s
(501-frame) window; audiosronnx slides it with a crossfade. Upstream's own
restore_inmemsegments long audio too. spis genuinely unused. The module signature isforward(sp, mel_orig), but zeroing or scalingspchanges the output by exactly zero, so the exported graph takes onlymel_orig. That is correct, not a folded constant.- Measure parity on speech. The same graph reads 6.8e-02 against random noise and 2.3e-04 against speech. Noise is out of domain for a speech restoration model and drives it into a regime where float error through a 70M-parameter ResUNet amplifies.
- Generative. It resynthesises rather than filters, so restored detail is invented.
License
MIT, inherited from the upstream VoiceFixer release.