{"id":34847556,"url":"https://github.com/hackjutsu/vibe-stt-server","last_synced_at":"2026-05-01T08:32:47.427Z","repository":{"id":327157553,"uuid":"1106795341","full_name":"hackjutsu/vibe-stt-server","owner":"hackjutsu","description":"Speech-to-Text transcription service backed by Whisper","archived":false,"fork":false,"pushed_at":"2025-12-02T03:18:07.000Z","size":12,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-12-04T20:20:47.720Z","etag":null,"topics":["chatbot","fastapi","whisper"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/hackjutsu.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-11-30T00:31:41.000Z","updated_at":"2025-12-02T03:18:10.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/hackjutsu/vibe-stt-server","commit_stats":null,"previous_names":["hackjutsu/vibe-stt-server"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/hackjutsu/vibe-stt-server","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hackjutsu%2Fvibe-stt-server","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hackjutsu%2Fvibe-stt-server/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hackjutsu%2Fvibe-stt-server/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hackjutsu%2Fvibe-stt-server/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hackjutsu","download_url":"https://codeload.github.com/hackjutsu/vibe-stt-server/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hackjutsu%2Fvibe-stt-server/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32490810,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-30T13:12:12.517Z","status":"online","status_checked_at":"2026-05-01T02:00:05.856Z","response_time":64,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["chatbot","fastapi","whisper"],"created_at":"2025-12-25T18:52:03.379Z","updated_at":"2026-05-01T08:32:47.416Z","avatar_url":"https://github.com/hackjutsu.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Vibe STT Server\n\n[GitHub Repository](https://github.com/hackjutsu/vibe-stt-server).\n\nFastAPI-based one-shot Whisper transcription service following the [design](./discussions/init_design.md). This is a project for near real-time speech-to-text transcription in headless Ubuntu machine(with Nvidia GPU) to empower [the local voice chatbot project](https://github.com/hackjutsu/vibe-speech). When running in a more powerful machine, the speech-to-text latency is reduced from seconds to \u003c0.5s with the model `large-v3-turbo` compared to running in the local dev machine.\n\n## Endpoints\n- `GET /health` → `{ \"status\": \"ok\" }`\n- `GET /info` → model/device/config info\n- `POST /transcribe` → transcribe a single audio session\n\n### POST /transcribe body\n```json\n{\n  \"audio\": \"\u003cbase64 mono 16 kHz int16 PCM\u003e\",\n  \"sample_rate\": 16000,\n  \"language\": \"en\",\n  \"initial_prompt\": \"optional\",\n  \"beam_size\": 5\n}\n```\n\nResponse:\n```json\n{\n  \"text\": \"...\",\n  \"info\": {\n    \"language\": \"en\",\n    \"duration_ms\": 2000,\n    \"decode_time_ms\": 120,\n    \"num_samples\": 32000\n  }\n}\n```\n- `language` is optional; when omitted, Whisper auto-detects the language.\n- `beam_size` defaults to the server setting if not provided.\n\n### GET /info response\n```json\n{\n  \"model\": \"large-v3-turbo\",\n  \"device\": \"auto\",\n  \"compute_type\": \"auto\",\n  \"sample_rate\": 16000,\n  \"num_workers\": 1,\n  \"cpu_threads\": 0,\n  \"default_beam_size\": 5\n}\n```\n\n## Config (env vars)\n- `WHISPER_MODEL` (default: `large-v3-turbo`)\n- `WHISPER_DEVICE` (default: `auto`) — e.g., `cpu`, `cuda`\n- `WHISPER_COMPUTE_TYPE` (default: `auto`) — e.g., `float16`, `int8_float16`\n- `WHISPER_CPU_THREADS` (default: `0` for library default)\n- `WHISPER_NUM_WORKERS` (default: `1`)\n- `WHISPER_BEAM_SIZE` (default: `5`)\n- `HOST` (default: `0.0.0.0`)\n- `PORT` (default: `8000`)\n- `LOG_LEVEL` (default: `INFO`)\n\n## Run \n\n### macOS dev, CPU\n```bash\npython -m venv .venv\nsource .venv/bin/activate\npip install -r requirements.txt\npython -m app.main\nuvicorn app.main:app --reload --host 0.0.0.0 --port 8000  # for local dev hot-reload\n```\n\n### Linux with GPU\nCheck out this [summary](./discussions/resolve_whisper_linux_gpu_deps.md) on how to resolve dependencies issue with running Nvidia GPU with Ubuntu.\n\nFor local development.\n```bash\npython -m venv .venv\nsource .venv/bin/activate\npip install -r requirements.txt\npython -m app.main\n```\n\nFor running service as a detached tmux session\n```bash\ntmux new -s stt-session -d 'uvicorn app.main:app --host 0.0.0.0 --port 8000'\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhackjutsu%2Fvibe-stt-server","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhackjutsu%2Fvibe-stt-server","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhackjutsu%2Fvibe-stt-server/lists"}