# datascale-ai/opentalking

OpenTalking: An industrial-grade open-source AI digital human framework that supports real-time conversation, private deployment, and pluggable models.

Repository: https://github.com/datascale-ai/opentalking
Canonical: https://ross.abutalabs.com/products/opentalking
Homepage: https://www.opentalking.net/
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-14T08:10:39+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 57, longevity 9
- inputs: {"age_days": 139, "days_push": 19, "days_rel": 79, "gap_med": null, "n_releases_24m": 1}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2897, forks 580 (observed 2026-08-28T04:07:28.698163+00:00)

## What it is
OpenTalking is an open-source Python framework for building real-time AI digital-human (talking avatar) conversation products. It orchestrates a full pipeline including LLM, STT, TTS, interruption control, subtitles, WebRTC audio/video playback, and pluggable model backends like Wav2Lip, FlashTalk, and FasterLivePortrait.

## Use cases
- build a real-time talking avatar chatbot
- deploy a private digital human assistant
- stream an LLM-driven avatar over WebRTC
- create lip-synced talking head videos from audio
- clone a digital human from an uploaded video
- run a self-hosted voice conversation agent with TTS and STT

## When to choose
- you need an end-to-end, self-hosted digital-human conversation stack
- you want pluggable LLM/STT/TTS and lip-sync model backends
- you need real-time WebRTC streaming with interruption handling
- you want video creation or avatar cloning workflows included

## When to avoid
- you only need a simple text chatbot without an avatar
- you need a fully managed cloud service rather than self-hosting
- you lack GPU resources for the video model backends
- you need a lightweight avatar renderer rather than a full pipeline

## Facets
- artifact type: framework
- maturity: active
- function: llm-inference, tts, speech-recognition, rag, chatbot, web-framework, websocket, video-processing, audio-processing, agent-framework
- domain: artificial-intelligence, large-language-models, chatbots, speech-processing, web-development, self-hosted
- platform: python, self-hosted, cross-platform
- tags: digital-human, avatar, webrtc, talking-head, lip-sync, wav2lip, liveportrait, real-time-conversation, private-deployment, virtual-assistant, real-time, docker, web-server, gpu

## Member repositories
- datascale-ai/opentalking (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.698163+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 2026-08-30T07:35:14.064594+00:00, confidence not recorded.
  - readme: https://github.com/datascale-ai/opentalking (fetched 2026-08-28T04:07:28.698163+00:00, sha bf15b2eb35ed)
  - homepage: https://www.opentalking.net/ (fetched 2026-08-29T09:50:02.930050+00:00, sha e99cd5c433e7)
- Data as of 2026-08-30T08:39:29.467469+00:00.