jishengpeng/WavTokenizer
[ICLR 2025] SOTA discrete acoustic codec models with 40/75 tokens per second for audio language modeling
This tool helps researchers and AI developers working with audio to convert raw speech, music, or other sounds into a highly compressed sequence of 'tokens.' These tokens are a simplified representation of the audio, making it much easier and faster to process within advanced AI systems like audio language models. It takes audio files as input and outputs these discrete audio tokens, or can reconstruct audio from previously generated tokens.
1,279 stars. No commits in the last 6 months.
Use this if you need to efficiently represent audio with very few data points per second (40-75 tokens/second) while maintaining high sound quality for applications like audio language modeling or generative AI for sound.
Not ideal if your primary goal is basic audio transcription or simple signal processing tasks that don't require advanced discrete representation for AI models.
Stars
1,279
Forks
111
Language
Python
License
MIT
Category
Last pushed
Mar 02, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/jishengpeng/WavTokenizer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
kan-bayashi/ParallelWaveGAN
Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN & HiFi-GAN & StyleMelGAN) with Pytorch
fatchord/WaveRNN
WaveRNN Vocoder + TTS
shangeth/wavencoder
WavEncoder is a Python library for encoding audio signals, transforms for audio augmentation,...
rishikksh20/iSTFTNet-pytorch
iSTFTNet : Fast and Lightweight Mel-spectrogram Vocoder Incorporating Inverse Short-time Fourier...
seungwonpark/melgan
MelGAN vocoder (compatible with NVIDIA/tacotron2)