Ultrafast serverless GPU inference, sandboxes, and background jobs

C inference for Qwen3-ASR 0.6b and 1.7b transcriptions models

Two-stage Dereverberation Algorithm using DNN-supported multi-channel linear filtering and single-channel non-linear post-filtering

Python scraper based on AI

Real-time speech enhancement in the browser using pure C + WASM SIMD 128-bit.

Speed-optimized streaming neural speech enhancement network

Score-based Generative Models (Diffusion Models) for Speech Enhancement and Dereverberation

AirLLM 70B inference with single 4GB GPU

Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice…

Pytorch implementation of Avatar Forcing: Real-Time Interactive Head Avatar Generation for Natural Conversation

Simultaneous speech-to-text model

Fast and local neural text-to-speech engine

Fast audio super resolution from 16khz to 48khz.

MNN is a blazing fast, lightweight deep learning framework, battle-tested by business-critical use cases in Alibaba. Full multimodal LLM Android App:[MNN-LLM-Android](./apps/Android/MnnLlmChat/READ…

SQL Native Memory Layer for LLMs, AI Agents & Multi-Agent Systems

🍁 Sycamore is an LLM-powered search and analytics platform for unstructured data.

Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.

dLLM: Simple Diffusion Language Modeling

Optimized Whisper models for streaming and on-device use

기타 포지션별 코드와 스케일을 쉽게 찾을 수 있는 앱

team ORI of the 1st Hackathon Contest of Hyupseongdae

🚀 Web scraping for humans

Omnilingual ASR Open-Source Multilingual SpeechRecognition for 1600+ Languages

This is a repository that collects common audio noise reduction models, using Gradio to demonstrate the use of each model, which is very friendly for beginners.

An AI-Powered Speech Processing Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Enhancement, Separation, and Target Speaker Extraction, etc.

finetune llm part for spark-tts model

TIGER: Time-frequency Interleaved Gain Extraction and Reconstruction for Efficient Speech Separation

Easy to use stem (e.g. instrumental/vocals) separation from CLI or as a python package, using a variety of amazing pre-trained models (primarily from UVR)