A list of all public EEG-datasets

Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation

SpeechJudge: Towards Human-Level Judgment for Speech Naturalness (https://arxiv.org/abs/2511.07931)

[NeurIPS' 25] Benchmark for evaluating TTS models on complex prosodic, expressiveness, and linguistic challenges.

[ACL 2024] Official PyTorch code for extracting features and training downstream models with emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representation

Sync papers from Zotero to a reMarkable tablet

A curated list of projects related to the reMarkable tablet

A Quirky Assortment of CuTe Kernels

Physics of Language Models: Part 4.2, Canon Layers at Scale where Synthetic Pretraining Resonates in Reality

PyTorch implementation of the Mamba-3 architecture

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding

FACodec: Speech Codec with Attribute Factorization used for NaturalSpeech 3

DACVAE

Kernels, of the mega variety :)

Interactive visualizations of the geometric intuition behind diffusion models.

🚀 Efficient implementations for emerging model architectures

A torch implementation of a recursion which turns out to be useful for RNN-T.

Open-source release accompanying Gao et al. 2025

GPU kernels for state space models

Jax Codebase for Evolutionary Strategies at the Hyperscale

On-device TTS model by Neuphonic

Python library for backtesting trading strategies & analyzing financial markets (formerly pythalesians)

Omnilingual ASR Open-Source Multilingual SpeechRecognition for 1600+ Languages

Text to speech alignment using CTC forced alignment

LongCat Audio Tokenizer and Detokenizer

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.