# AutoArk/GPA

[AutoArk] GPA (General Purpose Audio) can do ASR, TTS and voice conversion with one tiny model!

Repository: https://github.com/AutoArk/GPA
Canonical: https://ross.abutalabs.com/products/gpa
Homepage: https://autoark.github.io/GPA/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: automatic-speech-recognition, text-to-speech, transformer, voice-conversion, asr, tts, vc
Last push: 2026-05-25T08:02:41+00:00

## Health v2 (maintenance only)
Score: 54/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 84, release rhythm 35, longevity 18
- inputs: {"age_days": 260, "days_push": 100, "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 2762, forks 186 (observed 2026-08-28T04:07:18.065142+00:00)

## What it is
GPA (General Purpose Audio) is a unified autoregressive audio-language model that performs text-to-speech, automatic speech recognition, and voice conversion within a single small model. It offers native PyTorch/Hugging Face workflows plus an ONNX runtime with CLI, FastAPI service, and browser UI deployment options.

## Use cases
- transcribe speech to text with a tiny model
- synthesize speech from text with voice cloning
- convert one voice to another
- deploy speech models on edge devices with ONNX
- run a self-hosted speech API service
- fine-tune a unified audio model on custom data

## When to choose
- you want one compact model covering ASR, TTS, and voice conversion instead of separate pipelines
- you need lightweight, quantized speech runtimes for edge or CPU deployment
- you want an open Apache-2.0 speech model with ONNX, FastAPI, and browser UI options

## When to avoid
- you need state-of-the-art accuracy on a single specific speech task and can afford larger specialized models
- you need low-latency streaming or production-grade enterprise speech services
- you require extensive language coverage beyond what the model supports

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, tts, machine-learning, llm-inference, audio-processing
- domain: speech-processing, artificial-intelligence, deep-learning
- platform: python, cross-platform, cli
- tags: audio-language-model, voice-conversion, onnx-runtime, unified-model, speech-synthesis, edge-deployment, quantization, audio, gpu, web-server

## Member repositories
- AutoArk/GPA (main) score 54

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:18.065142+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 2026-08-30T08:17:26.893213+00:00, confidence not recorded.
  - readme: https://github.com/AutoArk/GPA (fetched 2026-08-28T04:07:18.065142+00:00, sha a78db79c457c)
  - homepage: https://autoark.github.io/GPA/ (fetched 2026-08-29T09:56:29.043778+00:00, sha cd628e62bfd0)
- Data as of 2026-08-30T08:39:29.467469+00:00.