Ultra-low bitrate speech codec (0.27-1 kbps) with cross-modal alignment and real-time capabilities
Ultra-low-bitrate Speech Codec for Speech Language Modeling Applications
ICASSP 2024 - Generative De-Quantization for Neural Speech Codec via Latent Diffusion.
microphone array speech generator (MASG) in room acoustic
AudioCodec-Hub is a Python library for encoding and decoding audio data, supporting various neural audio codec models
3GPP Codec for Enhanced Voice Services (EVS)
Official repo of ICASSP 2021 paper Source-Aware Neural Speech Coding for Noisy Speech Compression (SANAC)
Official inference code and checkpoints for CodecSlime.
Official PyTorch implementation of GS-Codec: A Gaussian-Splatting Bottleneck for Neural Audio Coding (NeurIPS 2026)
Pure-Rust Opus audio codec (RFC 6716) with Ogg encapsulation (RFC 7845). Encoder and decoder for SILK, CELT and hybrid modes. No C, no FFI, no dependencies.
Low-bitrate factorized speech codec on top of FACodec: FSQ distillation to replace RVQ (WER, mel and speaker similarity vs bitrate, with bootstrap CIs) and a learned reverberation/noise "environment" factor (partial result, with limitations).
CPNT2 超低码率语音编解码 Android Demo：BLE 双机通话，0.15–3.2 kbps，Codec2 + ONNX 神经编解码
Experimental Lyra V2 speech codec for WebGPU/WGSL with a browser recording demo at 3.2, 6, and 9.2 kbps.
1.09 kbps neural speech codec for speech language models and TTS. Two FSQ token streams, 24 kHz output. Strongest on Indian languages.
Hide digital data inside speech-shaped audio that survives Zoom, Discord, WhatsApp, and cellular voice. Reproducible Pareto curve of six trained codecs spanning 76 bps (cellular) to 3196 bps (Zoom-class) with listenable demos.
To associate your repository with the speech-codec topic, visit your repo's landing page and select "manage topics."