Steven Zheng

smol-audio. Practical, Colab-friendly notebooks for fine-tuning and running audio AI models

Finetune-Voxtral-ASR. Python

nano-cohere-transcribe. Pure-PyTorch inference for CohereLabs/cohere-transcribe-03-2026 (2B Conformer + Transformer ASR, 14 languages).

Finetune-Parakeet. Python

Finetune-Dia-TTS. Python

Llasa-GRPO. Python

Finetune-AudioFlamingo3. Python

LLaSA_training. LLaSA: Scaling Train-time and Test-time Compute for LLaMA-based Speech Synthesis

Learning-Audio-ML.

SmolVoice. Python

finetune-gemma-audio. Python

notebooks. Jupyter Notebook

TTS-Audio-Suite. A ComfyUI custom node integration for multi-language High-quality Text-to-Speech and Voice Conversion nodes using multiple engines like RVC, ResembleAI's Chatterbox TTS, F5-TTS, Higgs Audio 2 and Microsoft VibeVoice with unlimited text length, SRT timing, Character support, Audio Analyzer, Silent Speech Analyzer, audio edit and more!!

ddp_from_scratch. Python

CoVR. Official PyTorch implementation of the paper "CoVR: Learning Composed Video Retrieval from Web Video Captions".

genir-search-reco. Internship project at Criteo AI Lab

Ming-UniAudio. Python

clawdbot. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞