Open-Source Frontier Voice AI

Versatile Evaluation of Speech and Audio

A toolkit dedicate for speech evaluation.

Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM

The official Meta Llama 3 GitHub site

List of Dirty, Naughty, Obscene, and Otherwise Bad Words

Audio Codec Speech processing Universal PERformance Benchmark

State-of-the-art audio codec with 90x compression factor. Supports 44.1kHz, 24kHz, and 16kHz mono/stereo audio.

The official repository of Dynamic-SUPERB.

Robust Speech Recognition via Large-Scale Weak Supervision

Large, modern dataset for speech recognition

E2E system with LF-MMI; word N-gram for Mandarin

End-to-End Speech Processing Toolkit

kaldi-asr/kaldi is the official location of the Kaldi project.

FSA/FST algorithms, differentiable, with PyTorch compatibility.

Script to perform statistical significance test between ASR hypotheses.