每日学习笔记

Voice Activity Detector (VAD) : low-latency, high-performance and lightweight

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

Official Implementation of GLAP - General Language Audio Pretraining

VoiceBench: Benchmarking LLM-Based Voice Assistants

Official inference framework for 1-bit LLMs

Open-source industrial-grade ASR models supporting Mandarin, Chinese dialects and English, achieving a new SOTA on public Mandarin ASR benchmarks, while also offering outstanding singing lyrics rec…

https://hf.co/hexgrad/Kokoro-82M

pHash - the open source perceptual hash library

partitioned block based frequency domain Kalman filter

TTSAudioNormalizer is a specialized tool for TTS data production, featuring descriptive statistical analysis of audio loudness and loudness normalization operations.

Different implementations of "Weighted Prediction Error" for speech dereverberation

KWS demo based on CTC prefix beam search.

Python library & examples for Masked Language Model Scoring (ACL 2020)

使用Bert，ERNIE，进行中文文本分类

Experiments with BitNet inference on CPU

大规模中文自然语言处理语料 Large Scale Chinese Corpus for NLP

中文标点符号模型，可以给文本添加标点符号。

Audio Editor

Official implementation of "Separate Anything You Describe"

宽带波束形成，鄢社峰优化波束书本复现代码（第八章）

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

Pytorch implementation of BiFSMNv2, TNNLS 2023

ManyEars Sound Source Localization, Tracking and Separation

End-to-end ASR/LM implementation with PyTorch