ASLP-lab

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

OSUM. OSUM & OSUM-EChat, open speech understanding model and empathetic spoken chatbot based on it, open-sourced by ASLP@NPU.

WenetSpeech-Yue. A Large-scale Cantonese Speech Corpus with Multi-dimensional Annotation

SongEval. A song aesthetic evaluation toolkit trained on SongEval.

MeanVC. A Lightweight and Streaming Zero-Shot Voice Conversion via Mean Flows

VoiceSculptor. An instruct text-to-speech solution based on LLaSA and CosyVoice2 developed by the ASLP lab and collaborators.

WenetSpeech-Chuan. Official repository for the WenetSpeech-Chuan dataset.

DiffRhythm2. Di♪♪Rhythm 2: Efficient And High Fidelity Song Generation Via Block Flow Matching

WenetSpeech-Wu-Repo. A Large-scale Wu Dialect Speech Corpus with Multi-dimensional Annotations

SongFormer. Python

Easy-Turn. Open-Source Turn-Taking Detection Model and Dataset for Full-Duplex Spoken Dialogue Systems

SenSE. Official code of SenSE.

Speaker-Reasoner. Speaker-Reasoner: Scaling Interaction Turns and Reasoning Patterns for Timestamped Speaker-Attributed ASR

YingMusic-Singer-Plus. YingMusic-Singer-Plus: Controllable Singing Voice Synthesis with Flexible Lyric Manipulation and Annotation-free Melody Guidance

LLaSA_Plus. Llasa Speed Up

Hum-Dial. ICASSP2026 HumDial Challenge

FlashTTS. Fast Streaming TTS with MTP Acceleration and X-pred Mean Flow Distillation

MINT-Bench. Python

LLaSE-G1. LLaSE-G1: Incentivizing Generalization Capability for LLaMA-based Speech Enhancement

OmniCodec. OmniCodec: Low Frame Rate Universal Audio Codec with Semantic–Acoustic Disentanglement

HumDial-FDBench. The Full-Duplex Interaction Track of the ICASSP 2026 Human-like Spoken Dialogue Systems Challenge aims to advance the evaluation of full-duplex dialogue systems by in- troducing a dual-channel dialogue dataset of real human- recorded conversations.

OSUM-Pangu. An Open-Source Multidimension Speech Understanding Foundation Model Built upon OpenPangu on Ascend NPUs

FastTurn.

ArxivWatcher. Python

FMSU. Towards Fine-Grained Multi-Dimensional Speech Understanding: Data Pipeline, Benchmark, and Model

M7-TTS. M7-TTS: A Mini-Scale Multilingual and Multi-Dialect Text-to-Speech Language Model with Mimi codec and Multi Token Prediction

MeanVC2.

Music-Semantic-VAE.

SmartGlasses. This challenge focuses on evaluating speech recognition and semantic understanding capabilities of AI glasses in complex real-world environments.

C2SER. We propose C2SER, a novel audio-language model designed to enhance the stability and accuracy of speech emotion recognition through contextual perception and chain of Thought (CoT).

Smart-Glass-Challenge. Shell

Automatic-Song-Aesthetics-Evaluation-Challenge. TypeScript

MSU-Bench. Open repository of "MSU-Bench: Towards Understanding the Conversational Multi-Speaker Scenarios"

HumDial-Challenge. TypeScript

Stream-TN. HTML

MINT-Bench-Demo. Demo page of MINT-Bench

HumDial-EIBench. Python

Llasa-1B-Yue-Updated. Python

gen-se. GenSE: Generative Speech Enhancement via Language Models using Hierarchical Modeling

Speaker-Reasoner-Demo. HTML

SongEval_anonymous. Python

DiffRhythm2.github.io. JavaScript

YingMusic-Singer-Plus-Demo. HTML

ChinaVoices-Challenge. HTML

osum-echat.github.io. OSUM-EChat demo page

DiffRhythm.github.io. HTML

MLC-SLM-Baseline. The project is associated with the recently-launched INTERSPEECH 2025 Workshop on Multilingual Conversational Speech Language Model (MLC-SLM) to provide participants with baseline systems for speech recognition and speaker diarization in multilingual conversational scenario.

OSUM.github.io. HTML

msu-bench.github.io. Online demo page for "MSU-Bench: Towards Understanding the Conversational Multi-Speaker Scenarios"