{"data":{"slug":"autoark-gpa","name":"GPA","tagline":"[AutoArk] GPA (General Purpose Audio) can do ASR, TTS and voice conversion with one tiny model!","github_url":"https://github.com/AutoArk/GPA","owner":"AutoArk","repo":"GPA","owner_avatar_url":"https://avatars.githubusercontent.com/u/224494850?v=4","primary_language":"Python","stars":1152,"forks":119,"topics":["asr","automatic-speech-recognition","text-to-speech","transformer","tts","vc","voice-conversion"],"archived":false,"github_pushed_at":"2026-05-25T08:02:41+00:00","maintenance_label":"Steady","url":"https://www.graphcanon.com/tools/autoark-gpa","markdown_url":"https://www.graphcanon.com/tools/autoark-gpa.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/autoark-gpa","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=autoark-gpa","description":"[AutoArk] GPA (General Purpose Audio) can do ASR, TTS and voice conversion with one tiny model!","homepage_url":"https://autoark.github.io/GPA/","license":"Apache-2.0","open_issues":4,"watchers":124,"ai_summary":null,"readme_excerpt":"<div align=\"center\">\n  <img src=\"scripts/GPA.png\" width=\"80%\" alt=\"GPA Logo\"/>\n\n# GPA: One Model for Speech Recognition, Text-to-Speech, and Voice Conversion\n\n\n    \n\n</div>\n\n<details open>\n<summary><strong>📢 Announcements</strong></summary>\n\n<div style=\"max-height: 150px; overflow-y: auto; border: 1px solid #ddd; padding: 10px; margin-top: 8px;\">\n\n- 🚀 **2026.04.29: [GPA-v1.5](GPA_1.5/) is here!**\n  GPA v1.5 Delivers near-SOTA TTS and ASR performance—in a single unified model. [Start here →](GPA_1.5/README.md)\n\n- 🚀 **2026.04.29: GPA-v1.5 ONNX Runtime is now available!**\n  Run ASR/TTS through ONNX CLI tools, a FastAPI service, or the browser UI with the new **[GPA-v1.5 ONNX runtime guide](GPA_1.5/onnx_runtime/README.md)** and **[runtime asset bundle](https://huggingface.co/AutoArk-AI/GPA-v1.5-onnx-runtime)**.\n\n- 🆕 **2026.04.07: [GPA-TTS FP16/FP32 Decoder](GPA_TTS/) — Higher-quality decoder options now available!**\n  For users with extra compute headroom, FP16 and FP32 SparkDetokenizer decoders are now available alongside INT8, delivering more stable and higher-quality speech synthesis. Selectable at runtime via CLI, API, or Web UI. [Details →](GPA_TTS/README.md)\n\n- 📌 **2026.03.31: [GPA-TTS](GPA_TTS/) — Standalone lightweight TTS runtime released!**\n  Extracted from GPA with INT8/INT4 quantization for edge deployment. Among the smallest open-source TTS runtimes with voice cloning support! [Details →](GPA_TTS/README.md)\n\n- 📚 **GPA-v1.0 docs have moved.**\n  The original GPA-0.3B-preview quick start, deployment, benchmark, and evaluation pages now live in **[docs/GPA-v1.0.md](docs/GPA-v1.0.md)**.\n\n</div>\n\n</details>\n\n<br>\n\n<div align=\"center\" style=\"margin: 22px 0 28px;\">\n  <video src=\"https://github.com/user-attachments/assets/dbf223ef-4b64-40ca-bb50-4f3541ec0ffb\" width=\"85%\" controls loop playsinline style=\"border: 1px solid #e5e7eb; border-radius: 8px;\"></video>\n  <br>\n  <sub>All in one, built for all.</sub>\n  <br>\n  <sub>A single model delivering near-SOTA performance on TTS and ASR — fully unified, fully open!</sub>\n</div>\n\n<br>\n\n## 📖 Abstract\n\n**GPA** stands for **General Purpose Audio**.\n\nA student’s GPA unifies performance across diverse subjects—from Calculus to Gym—into a single metric. Likewise, our GPA model integrates the three core audio tasks—TTS, ASR, and Voice Conversion—into one auto-regressive transformer.\n\nGPA-v1.5 now delivers [near-SOTA performance on ASR and TTS](#-gpa-v15-evaluation-metric-results) in a single unified model, with VC support on the roadmap.\n\n<br>\n\n<div align=\"center\" style=\"margin: 20px 0 24px;\">\n  <img src=\"docs/GPA_v1.5.jpeg\" width=\"86%\" alt=\"GPA Unified Speech Model Overview\" style=\"border: 1px solid #e5e7eb; border-radius: 8px;\"/>\n  <br>\n  <sub><strong>Figure 1.</strong> GPA unifies speech understanding and generation in a single autoregressive audio-language model.</sub>\n</div>\n\n<br>\n\n<div align=\"center\">\n\n[🗺️ Roadmap](#-roadmap) · [🚀 GPA-v1.5 Release](#-gpa-v15-release) · [🎙️ GPA-TTS](#-gpa-tts-edge-ready-voice-cloning-tts) · [🧭 GPA-v1.0 Archive](#-gpa-v10-archive) · [📊 GPA-v1.5 Evaluation](#-evaluation-metric-results) · [🔗 Citation](#-citation)\n\n</div>\n\n## 🗺️ Roadmap\n\n| Category | Item | Status |\n| :--- | :--- | :---: |\n| **Core Features** | Unified LLM-based audio generation & understanding | ✅ |\n| | Native GPA-v1.5 Inference Pipeline | ✅ |\n| | Native GPA-v1.5 Training Pipeline | ✅ |\n| | GPA-v1.5 ONNX Runtime CLI/API/UI | ✅ |\n| | GPA-v1.5 Interactive Demo | ⬜ |\n| | GPA-v1.5 Basic Service Deployment (vLLM/FastAPI) | ⬜ |\n| | Paper (ArXiv) | ✅ |\n| **Model Releases** | **GPA-0.3B-preview** | ✅ |\n| | **[GPA-v1.5](GPA_1.5/)** — major mainline release | ✅ |\n| | **[GPA-TTS](GPA_TTS/)** — Lightweight TTS runtime (INT8/FP16/FP32 + INT4 ONNX) | ✅ |\n| **GPA-v1.5 Next Steps** | Voice Conversion native path | ⬜ |\n| | Expanded deployment recipes | ⬜ |\n| **Frameworks** | torch | ✅ |\n| | vllm | ✅ |\n| | llama-cpp | ✅ |\n| | sglang | ✅ |\n| | mlx-lm | ✅ |\n| | rknn | ⬜ |\n\n## 🚀 GPA-v1.5 Releas","github_created_at":"2025-12-16T05:20:34+00:00","created_at":"2026-07-11T12:13:07.173454+00:00","updated_at":"2026-07-11T12:13:30.067027+00:00","categories":[{"slug":"vector-databases","name":"Vector Databases","url":"https://www.graphcanon.com/categories/vector-databases","markdown_url":"https://www.graphcanon.com/categories/vector-databases.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/vector-databases"},{"slug":"llm-frameworks","name":"LLM Frameworks","url":"https://www.graphcanon.com/categories/llm-frameworks","markdown_url":"https://www.graphcanon.com/categories/llm-frameworks.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/llm-frameworks"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"voice-conversion","name":"voice-conversion"},{"slug":"automatic-speech-recognition","name":"automatic-speech-recognition"},{"slug":"asr","name":"asr"},{"slug":"text-to-speech","name":"text-to-speech"},{"slug":"python","name":"python"},{"slug":"tts","name":"tts"},{"slug":"vc","name":"vc"},{"slug":"transformer","name":"transformer"}],"trust":{"provenance":{"is_fork":false,"github_id":1117288212,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-11T12:13:07.837Z","maintenance":{"label":"Steady","score":60,"methodology":"github_public_v1","releases_90d":0,"days_since_push":47,"last_release_at":null},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":34,"high_count":0,"last_scan_at":"2026-07-11T12:13:16.550Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-07-11T12:13:16.240Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-07-11T12:13:16.240Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-07-11T12:13:16.240Z"}}}}