# AIGC-Audio/AudioGPT

AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Head

Repository: https://github.com/AIGC-Audio/AudioGPT
Canonical: https://ross.abutalabs.com/products/audiogpt
Homepage: https://huggingface.co/spaces/AIGC-Audio/AudioGPT
Language: Python
License: NOASSERTION
License Family: other
Topics: audio, gpt, music, sound, speech, talking-head
Last push: 2024-07-06T21:35:18+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 90
- inputs: {"age_days": 1266, "days_push": 788, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 10167, forks 850 (observed 2026-08-28T04:10:39.732404+00:00)

## What it is
AudioGPT is a Python framework that wraps multiple audio foundation models (for speech, singing, sound, and talking-head tasks) behind a GPT-style natural language interface. It supports tasks like text-to-speech, speech recognition, text-to-audio, and talking-head generation via pretrained models.

## Use cases
- generate speech from text
- convert speech to text
- generate music or singing from text
- create audio from text descriptions
- generate a talking head video
- enhance or separate speech audio

## When to choose
- you want a unified interface to many audio foundation models
- you need both speech understanding and audio generation in one tool
- you are reproducing the AudioGPT research paper

## When to avoid
- you need a production-ready, actively maintained audio API
- you want a simple single-task TTS or ASR tool
- you cannot run GPU-heavy models locally

## Facets
- artifact type: framework
- maturity: maintenance
- function: audio-processing, speech-recognition, tts, machine-learning, llm-inference
- domain: speech-processing, artificial-intelligence, large-language-models, machine-learning
- platform: python, cross-platform
- tags: audio-generation, talking-head, music-generation, foundation-models, research, audio, gpu

## Member repositories
- AIGC-Audio/AudioGPT (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:39.732404+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 2026-08-29T17:19:39.982254+00:00, confidence not recorded.
  - readme: https://github.com/AIGC-Audio/AudioGPT (fetched 2026-08-28T04:10:39.732404+00:00, sha a0de37714091)
  - homepage: https://huggingface.co/spaces/AIGC-Audio/AudioGPT (fetched 2026-08-29T08:19:13.373684+00:00, sha a2e3f72891ed)
- Data as of 2026-08-30T08:39:29.467469+00:00.