[ACL 2025 Main] UniCodec: a unified audio codec with a single codebook to support multi-domain audio data, including speech, music, and sound

Official github repo for TMMLU+, Large scale traditional chinese massive multitask language understanding

Ongoing research training transformer language models at scale, including: BERT & GPT-2

ACM MM 2021: 'Is Someone Speaking? Exploring Long-term Temporal Features for Audio-visual Active Speaker Detection'

A collection of prompts, system prompts and LLM instructions

使用繁體中文資料集做的 Embedding 模型評測

Transcription, forced alignment, and audio indexing with OpenAI's Whisper

Awesome speech/audio LLMs, representation learning, and codec models

Megatron-LM setup in the smol-cluster

CnOCR: Awesome Chinese/English OCR Python toolkits based on PyTorch. It comes with 20+ well-trained models for different application scenarios and can be used directly after installation. 【基于 PyTor…

🔊 Text-prompted Generative Audio Model - With the ability to clone voices

Python tools for processing the stackexchange data dumps into a text dataset for Language Models

A crawler for https://ndltd.ncl.edu.tw

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

PodcastMix A dataset for separating music and speech in podcasts.

RWKV is a RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, …

Python tools for processing the stackexchange data dumps into a text dataset for Language Models

Streamlit APPs that leverage the power of spaCy to assist language learning

結巴中文斷詞台灣繁體版本

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

StyleGAN3 + Inversion

Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

Open-source KVM software

Write a large text on your GitHub profile, with your commits history (contribution graph).

Pre-trained ELECTRA from Hong Kong data