AmphionTeam/FlexiCodec
[ICLR2026] FlexiCodec: A Dynamic Neural Audio Codec for Low Frame Rates
FlexiCodec helps developers working with speech and language models efficiently process audio. It takes an audio file as input and converts it into compressed, low-frame-rate semantic and acoustic codes, which can then be reconstructed back into high-quality audio or used in language models. This tool is ideal for engineers building or optimizing speech-based AI systems, particularly those focused on reducing computational costs.
Use this if you need to compress audio or convert it into very low-frame-rate tokens to reduce the computational load for speech language models or text-to-speech systems, while preserving high semantic quality.
Not ideal if your primary need is general-purpose audio compression for human listening rather than specialized processing for AI models, or if you require extremely high, fixed frame rates.
Stars
42
Forks
4
Language
Python
License
MIT
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/AmphionTeam/FlexiCodec"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
fatchord/WaveRNN
WaveRNN Vocoder + TTS
kan-bayashi/ParallelWaveGAN
Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN & HiFi-GAN & StyleMelGAN) with Pytorch
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)