# Henry-23/VideoChat

实时交互数字人，可自定义形象与音色，支持音色克隆，对话延迟低至3s。Real-time voice interactive digital human, customizable appearance and voice, supporting voice cloning, with initial package delay as low as 3s.

Repository: https://github.com/Henry-23/VideoChat
Canonical: https://ross.abutalabs.com/products/videochat
Language: Python
License: MIT
License Family: permissive
Topics: dialogue-systems, real-time, digital-human, gradio-python-app, lip-sync, musetalk, streaming, talking-head, asr, tts, end-to-end, multimodal-large-language-models
Last push: 2025-12-18T02:50:33+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 57, release rhythm 35, longevity 48
- inputs: {"age_days": 684, "days_push": 258, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1303, forks 172 (observed 2026-08-28T04:04:18.368186+00:00)

## What it is
A real-time voice-interactive digital human application that combines ASR, LLM, TTS, and talking-head generation (MuseTalk) into a low-latency conversational avatar. It supports customizable appearance and voice, including voice cloning, with first-packet latency as low as 3 seconds.

## Use cases
- build a real-time talking avatar that responds to voice
- create a custom digital human with my own face and voice
- clone a voice for an interactive virtual assistant
- stream a lip-synced talking head from an LLM chatbot
- run an end-to-end voice-to-avatar pipeline locally
- demo a low-latency conversational digital human

## When to choose
- you need a self-hosted real-time digital human with lip sync
- you want voice cloning and customizable avatar appearance
- you have a GPU (8GB+ for cascade, 20GB+ for end-to-end MLLM) and want low first-packet latency
- you want to swap in different ASR, LLM, or TTS components

## When to avoid
- you have no GPU or limited VRAM
- you need a production-grade hosted service rather than a demo app
- you need non-Linux deployment support
- you want a polished no-code product instead of a Python/Gradio pipeline

## Facets
- artifact type: application
- maturity: active
- function: speech-recognition, tts, chatbot, llm-inference, agent-framework, video-processing, audio-processing, machine-learning
- domain: artificial-intelligence, large-language-models, speech-processing, computer-vision, chatbots, deep-learning
- platform: python, self-hosted
- tags: digital-human, talking-head, lip-sync, voice-cloning, real-time, musetalk, gradio, asr, thg, multimodal, linux, gpu

## Member repositories
- Henry-23/VideoChat (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:18.368186+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:51:10.370792+00:00, confidence not recorded.
  - readme: https://github.com/Henry-23/VideoChat (fetched 2026-08-28T04:04:18.368186+00:00, sha ac1d32349940)
- Data as of 2026-08-30T08:39:29.467469+00:00.