{"id":44276495,"url":"https://github.com/potproject/uchinoko-studio","last_synced_at":"2026-02-10T20:36:19.590Z","repository":{"id":213901725,"uuid":"729005539","full_name":"potproject/uchinoko-studio","owner":"potproject","description":"Uchinoko Studio is a web application designed to facilitate real-time voice conversations with AI.","archived":false,"fork":false,"pushed_at":"2025-12-20T05:07:28.000Z","size":396,"stargazers_count":16,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-12-22T13:04:59.957Z","etag":null,"topics":["bert-vits2","claude","gpt-4","local-llm","style-bert-vits2","voicevox","web","whisper"],"latest_commit_sha":null,"homepage":"","language":"Svelte","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"agpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/potproject.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2023-12-08T07:27:05.000Z","updated_at":"2025-12-20T05:07:32.000Z","dependencies_parsed_at":"2023-12-25T09:20:34.590Z","dependency_job_id":"62d38a9c-fda0-4255-ac50-d9c9349613fa","html_url":"https://github.com/potproject/uchinoko-studio","commit_stats":null,"previous_names":["potproject/uchinoko-studio"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/potproject/uchinoko-studio","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/potproject%2Fuchinoko-studio","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/potproject%2Fuchinoko-studio/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/potproject%2Fuchinoko-studio/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/potproject%2Fuchinoko-studio/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/potproject","download_url":"https://codeload.github.com/potproject/uchinoko-studio/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/potproject%2Fuchinoko-studio/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29314890,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-10T17:48:59.043Z","status":"ssl_error","status_checked_at":"2026-02-10T17:45:37.240Z","response_time":65,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bert-vits2","claude","gpt-4","local-llm","style-bert-vits2","voicevox","web","whisper"],"created_at":"2026-02-10T20:36:17.748Z","updated_at":"2026-02-10T20:36:19.575Z","avatar_url":"https://github.com/potproject.png","language":"Svelte","funding_links":[],"categories":["Go"],"sub_categories":[],"readme":"# Uchinoko Studio (β)\n\u003cimg width=\"350\" alt=\"Uchinoko Studio (β)\" src=\"https://github.com/potproject/uchinoko-studio/assets/6498055/55dcb6d2-aff8-4bc3-9d0b-76bef1d88dae.png\"\u003e\n\n## Example (Ja)\n\nhttps://github.com/potproject/uchinoko-studio/assets/6498055/42a8e1fc-be28-4844-87ba-53bb6072ab71\n\nUsing OpenAI Transcriptions(Whisper), Style-Bert-VITS2, Anthropic Claude 3 Opus.\n\n__Multiple speakers \u0026 Real-Time Voice Conversation.__\n\nWARNING: This repository does not contain character images, LLM and Audio Model Data.\n\n## About\n\n__Work In Progress__\n\nUchinoko Studio is a web application designed to facilitate real-time voice conversations with AI. This is achieved by integrating various Large Language Models (LLMs), including Speech-To-Text LLMs like [Whisper](https://github.com/openai/whisper), Chat-based LLMs such as [GPT-4](https://openai.com/), and Text-To-Speech LLMs like [Bert-Vits2](https://github.com/fishaudio/Bert-VITS2).\n\n`Uchinoko` is a Japanese word that means `Waifu / My Daughter`.\n\nSee Article(Japanese Only): https://blog.potproject.net/2023/12/24/ai-web-uchinoko-studio/\n\n## Features\n\n* Real-time Voice Conversation\n* Multiple Speakers (Experimental)\n* Image Support (Experimental)\n* __Fast Response(Maybe 1 second or less)__\n* [Tailscale](https://tailscale.com/) Support\n* Run on Browser(Google Chrome Supported)\n* Mobile Browser Support(iOS Safari, Android Chrome)\n* Japanese Support(for now...)\n* Chat-based LLM Support: [OpenAI GPT(Cloud)](https://openai.com/gpt-4),[Anthropic Claude(Cloud)](https://www.anthropic.com/claude),[DeepSeek(Cloud)](https://www.deepseek.com),[Gemini(Cloud)](https://gemini.google.com),Local LLM(OpenAI `/v1/chat/completions` Compatible)\n* Speach-To-Text Support: [OpenAI Speech to Text API(Cloud)](https://platform.openai.com/docs/guides/speech-to-text),[Google Speech to Text API(Cloud)](https://cloud.google.com/speech-to-text),[Vosk Server(local)](https://github.com/alphacep/vosk-server),[SpeechRecognition(Web API)](https://developer.mozilla.org/docs/Web/API/SpeechRecognition)\n* Text-To-Speech Support: [Bert-Vits2(local)](https://github.com/fishaudio/Bert-VITS2), [Style-Bert-VITS2(local)](https://github.com/litagin02/Style-Bert-VITS2), [VOICEVOX(local)](https://voicevox.hiroshiba.jp/),[NijiVoice API(Cloud)](https://nijivoice.com), [Google Text To Speech API(Cloud)](https://cloud.google.com/text-to-speech),  [OpenAI Speech API(Cloud)](https://platform.openai.com/docs/guides/speech-to-text)\n* More bugs...\n\n## Getting Started\n\n__!WARNING! このアプリケーションはまだ開発中であり、正常に動作しない場合があります。__\n上手く動作しなくなった場合は、databaseフォルダを削除すると初期化されます。\n\nまた、アンチウイルスソフトウェアによっては、誤検知され、不正なファイルと判定されることがあるため、その場合は除外設定を行ってください。\n\n[Env Setting](#env-setting)を参考に.env.exampleから.envファイルを作成すれば実行できます。\n\n## Env Setting\n\n### Speech To Text\n\n* OpenAI Speech to Text API(Cloud): `OPENAI_SPEECH_TO_TEXT_API_KEY`\n* Google Speech to Text API(Cloud): `GOOGLE_SPEECH_TO_TEXT_API_KEY`\n* Vosk Server: `VOSK_SERVER_ENDPOINT`\n* SpeechRecognition(Web API): none\n\n### Chat-based LLM(Cloud)\n\n* OpenAI: `OPENAI_API_KEY`\n* Anthropic: `ANTHROPIC_API_KEY`\n* DeepSeek: `DEEPSEEK_API_KEY`\n* Gemini: `GEMINI_API_KEY`\n\n### Chat-based LLM(Local)\n\nLM StudioやOllamaのような、OpenAI互換のローカルLLMを使用することが出来ます。\n\n* `OPENAI_LOCAL_API_KEY` and `OPENAI_LOCAL_API_ENDPOINT`\n\n```\n# Example (Using LM Studio)\nOPENAI_LOCAL_API_KEY=\nOPENAI_LOCAL_API_ENDPOINT=http://localhost:1234/\n```\n\n```\n# Example (Using Ollama)\nOPENAI_LOCAL_API_KEY=\nOPENAI_LOCAL_API_ENDPOINT=http://localhost:11434/\n```\n\n### Text To Speech\n\nこのアプリケーションを使用する場合、以下のソフトウェアをローカルまたはネットワーク上で動作させておくことが前提です。\n\n* VOICEVOXの場合: `VOICEVOX_ENDPOINT`にVOICEVOX Engine APIのエンドポイントを設定してください。\n* BERTVITS2の場合: `BERTVITS2_ENDPOINT`にBert-VITS2 FastAPIのエンドポイントを設定してください。また、先にモデルのロードを行っていないと動作しません。\n* STYLEBERTVIT2の場合: `STYLEBERTVIT2_ENDPOINT`にStyle-Bert-VITS2 API Serverのエンドポイントを設定してください。モデルのロードは自動で行ってくれるため不要です。Bert-VITS2のAPIとの互換性はありません。\n* NijiVoice API(Cloud)の場合: `NIJIVOICE_API_KEY`を設定してください。\n* Google Text To Speech API(Cloud)の場合: `GOOGLE_TEXT_TO_SPEECH_API_KEY`を設定してください。\n* OpenAI Speech API(Cloud)の場合: `OPENAI_SPEECH_API_KEY`を設定してください。\n\n### Tailscale \n\n* Tailscaleを使用する場合、起動時にコンソールより認証URLが表示されるので、そこから認証を行ってください。Tailscaleのアカウントが必要です。\n* `TAILSCALE_ENABLED`を`true`に設定すると、[Tailscale](https://tailscale.com/)を使用してVPN上からアクセスできるようになります。\n  * これにより、自宅で起動して外からhttps通信で無いと動作しないSafariやiOSからもアクセスできるようになります。\n* `TAILSCALE_FUNNEL_ENABLED`を`true`に設定すると、\n[Tailscale Funnel](https://tailscale.com/kb/1223/funnel)機能を使用してパブリックアクセスできるようになります。何のことかわからなければ変更しないでください。\n\n## Development\n\n### Requirements\n\n* Go (Tested on 1.23.4/win-amd64)\n* Node.js (Tested on 20.11.1/win-amd64)\n* pnpm\n\n### Run on Local\n\n```\n## Easy Start (Windows)\ngit clone https://github.com/potproject/uchinoko-studio\ncd uchinoko-studio\ncopy server\\.env.example server\\.env\nrun-win.bat\n\n## Easy Start (Linux/mac)\ngit clone https://github.com/potproject/uchinoko-studio\ncd uchinoko-studio\ncp server/.env.example server/.env\nrun.sh\n```\n\n```bash\n# Install Dependencies\ncd front\npnpm install\npnpm build\n\n# Running\ncd ../server\ngo run main.go\n\n# 自動でブラウザが立ち上がります。\n# 立ち上がらない場合は、http://localhost:15000/ にアクセスしてください。\n# 話者(チャットプロンプト/使用するモデルの設定)などは、ブラウザより設定が可能です\n```\n\n## TODO\n\n* Docs\n* Frontend Design issue\n* Add Multilingual Support(EN and CH)\n* Fix Bugs...\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpotproject%2Fuchinoko-studio","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpotproject%2Fuchinoko-studio","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpotproject%2Fuchinoko-studio/lists"}