Di♪♪Rhythm: Blazingly Fast and Embarrassingly Simple End-to-End Full-Length Song Generation with Latent Diffusion

Spark-TTS Inference Code

FlashMLA: Efficient MLA decoding kernels

This is the official implementation of the LiSenNet

OSUM: Open Speech Understanding Model, open-sourced by ASLP@NPU.

Ultra-low bitrate neural audio codec (0.31~1.40 kbps) with a better semantic in the latent space.

Fast algorithm for determined blind source separation with update of demixing filters with joint adjustment of the remaining sources.

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

Target Speaker Extraction Toolkit

[Interspeech 2024] Hold Me Tight: Stable Encoder-Decoder Design for Speech Enhancement

Paderwasn is a collection of methods for acoustic signal processing in wireless acoustic sensor networks (WASNs).

A lightweight library for portable low-level GPU computation using WebGPU.

Reference implementation for DPO (Direct Preference Optimization)

10W首中文歌词数据库

Generate synthetic wind noise signals based on a wind speed profile.

LLM training in simple, raw C/CUDA

Stable Diffusion web UI

Synthesizes a room impulse response using a ray tracing simulation engine.

Graph Neural Networks for Sound Source Localization

Pitch detection and pitch tracking, voicing unvoicing detection (VAD)，基音检测

A python algorithm to change the pitch of the voice in real time

Unofficial pytorch reproduction for the paper "Utilizing Neural Transducers for Two-Stage Text-to-Speech via Semantic Token Prediction" (arXiv:2401.01498)

An optimized neural network operator library for chips base on Xuantie CPU.

" Music Style Transfer with Time-Varying Inversion of Diffusion Models"

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

Official implementation of Self-Remixing

多个SVC/TTS的C++推理库

中文人名语料库。人名生成器。中文姓名,姓氏,名字,称呼,日本人名,翻译人名,英文人名。可用于中文分词、人名实体识别。