AutoArk/GPA
[AutoArk] GPA (General Purpose Audio) can do ASR, TTS and voice conversion with one tiny 300M model!
Leverages a unified auto-regressive transformer backbone with shared semantic and acoustic modules to handle ASR, TTS, and voice conversion through task-specific input conditioning rather than separate models. Supports multiple inference frameworks (vLLM, llama-cpp, sglang, mlx-lm) and includes production-ready deployment via Docker/FastAPI, training pipelines with DeepSpeed, and interactive Hugging Face demos for experimentation.
Stars
97
Forks
16
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 04, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/AutoArk/GPA"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
dimonier/tg2obsidian
This bot pulls new messages from a Telegram chat or group and puts them into Obsidian vault on a...
Steve0929/tiktok-tts
Provides a simple way to generate text-to-speech audio files using TikTok's text-to-speech (TTS)...
Weilbyte/tiktok-tts
Generate TikTok Text-to-Speech voices in your browser
dngda/bot-whatsapp
Unmaintained - Multipurpose WhatsApp Bot 🤖 using open-wa/wa-automate-nodejs library! ✨
mark-rez/TikTok-Voice-TTS
Simple Python script to interact with the TikTok TTS Voices.