{"id":47783979,"url":"https://github.com/pyramidheadshark/milvm_stt","last_synced_at":"2026-04-03T14:04:38.189Z","repository":{"id":342551869,"uuid":"1164717078","full_name":"pyramidheadshark/milvm_stt","owner":"pyramidheadshark","description":"Windows tray app for instant voice transcription via OpenRouter + Gemini. No setup required.","archived":false,"fork":false,"pushed_at":"2026-03-16T13:26:18.000Z","size":319,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-03-16T16:15:16.050Z","etag":null,"topics":["desktop-app","fastapi","gemini","openrouter","pyinstaller","python","pywebview","speech-to-text","system-tray","voice-notes","voice-transcription","windows"],"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/pyramidheadshark.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":null,"dco":null,"cla":null}},"created_at":"2026-02-23T12:02:47.000Z","updated_at":"2026-03-16T13:26:22.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/pyramidheadshark/milvm_stt","commit_stats":null,"previous_names":["pyramidheadshark/milvm-stt"],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/pyramidheadshark/milvm_stt","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pyramidheadshark%2Fmilvm_stt","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pyramidheadshark%2Fmilvm_stt/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pyramidheadshark%2Fmilvm_stt/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pyramidheadshark%2Fmilvm_stt/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pyramidheadshark","download_url":"https://codeload.github.com/pyramidheadshark/milvm_stt/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pyramidheadshark%2Fmilvm_stt/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31355880,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-03T08:03:20.796Z","status":"ssl_error","status_checked_at":"2026-04-03T08:00:37.834Z","response_time":107,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 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":["desktop-app","fastapi","gemini","openrouter","pyinstaller","python","pywebview","speech-to-text","system-tray","voice-notes","voice-transcription","windows"],"created_at":"2026-04-03T14:04:37.414Z","updated_at":"2026-04-03T14:04:38.185Z","avatar_url":"https://github.com/pyramidheadshark.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Voice Transcriber\n\n[![CI](https://github.com/pyramidheadshark/milvm_stt/actions/workflows/ci.yml/badge.svg)](https://github.com/pyramidheadshark/milvm_stt/actions/workflows/ci.yml)\n[![codecov](https://codecov.io/gh/pyramidheadshark/milvm_stt/graph/badge.svg)](https://codecov.io/gh/pyramidheadshark/milvm_stt)\n[![Release](https://img.shields.io/github/v/release/pyramidheadshark/milvm_stt)](https://github.com/pyramidheadshark/milvm_stt/releases/latest)\n\nТрей-приложение для Windows: записал голосовую заметку — получил текст с заголовком через несколько секунд.\n\nРаботает на [OpenRouter](https://openrouter.ai/) + Google Gemini. Данные хранятся локально, никуда не отправляются кроме аудио на транскрибацию.\n\n---\n\n## Возможности\n\n- **Запись прямо в приложении** — через микрофон браузера, без сторонних программ\n- **Загрузка файла** — поддерживаются OGG, MP3, WAV, WebM, M4A, AAC, FLAC (до 25 МБ)\n- **Автоматический заголовок** — модель генерирует краткий заголовок на русском\n- **История и поиск** — все транскрибации сохраняются локально в SQLite, есть полнотекстовый поиск и пагинация\n- **Настройки в приложении** — API ключ и модель меняются без редактирования файлов\n- **Восстановление аудио** — если транскрибация упала, аудиофайл сохраняется и доступен для повторной попытки\n- **Трей** — приложение живёт в системном трее, окно открывается по клику\n\n---\n\n## Установка\n\n### Готовый .exe (Windows)\n\n1. Скачать `VoiceTranscriber.exe` из [Releases](https://github.com/pyramidheadshark/milvm_stt/releases/latest)\n2. Запустить — при первом запуске автоматически откроется панель настроек\n3. Ввести [API ключ OpenRouter](https://openrouter.ai/keys) и нажать Save\n\nБольше ничего не нужно. Данные и настройки хранятся в папке рядом с `.exe`.\n\n### Из исходников\n\nТребования: Python 3.11+, [uv](https://docs.astral.sh/uv/getting-started/installation/)\n\n```bash\ngit clone https://github.com/pyramidheadshark/milvm_stt.git\ncd milvm_stt\nuv sync\nuv run python tray.py\n```\n\n---\n\n## Настройки\n\nОткрыть панель настроек — кнопка **⚙** в заголовке окна.\n\n| Параметр | Описание |\n|---|---|\n| API Key | Ключ OpenRouter. Получить на [openrouter.ai/keys](https://openrouter.ai/keys) |\n| Model | ID модели OpenRouter с поддержкой аудио. По умолчанию — `google/gemini-2.5-flash-lite-preview-09-2025` |\n\nИзменения применяются сразу, без перезапуска.\n\nАльтернативно — через `.env` в папке с `.exe`:\n\n```env\nOPENROUTER_API_KEY=sk-or-v1-...\nMODEL=google/gemini-2.5-flash-lite-preview-09-2025\nPORT=8000\n```\n\nСписок моделей с поддержкой аудио: [openrouter.ai/models?input_modalities=audio](https://openrouter.ai/models?fmt=cards\u0026input_modalities=audio)\n\n---\n\n## Стоимость\n\nМодель по умолчанию: `google/gemini-2.5-flash-lite-preview-09-2025`\n\n| Тип токенов | Цена |\n|---|---|\n| Аудио | $0.30 / 1M |\n| Текст (выход) | $0.40 / 1M |\n\nЗаметка 1–2 минуты ≈ **$0.001–0.003**. При активном использовании (20 заметок/день) — около **$1–2 в месяц**.\n\n---\n\n## Разработка\n\n```bash\nuv sync\nmake dev          # веб-режим с hot-reload\nmake tray         # трей-приложение\nmake build        # собрать .exe\n```\n\nПолный список команд: `make help`\n\n### Тесты\n\n```bash\nuv run pytest --cov=. -q\n```\n\n81 тест, покрытие 95%. CI запускается на каждый push: ruff + mypy + pytest.\n\n### Релиз\n\n```bash\ngit tag v0.X.Y \u0026\u0026 git push origin v0.X.Y\n```\n\nGitHub Actions автоматически соберёт `.exe` и опубликует GitHub Release.\n\nПравила версий: `PATCH` — фиксы и мелкие улучшения, `MINOR` — новые фичи, `MAJOR` — breaking changes.\n\n---\n\n## Архитектура\n\n```\n├── tray.py              # точка входа — системный трей + окно pywebview\n├── main.py              # FastAPI: все эндпоинты\n├── config.py            # конфигурация, write_settings, reload_config\n├── paths.py             # разрешение путей (dev / PyInstaller bundle)\n├── services/\n│   ├── transcriber.py   # OpenRouter API, retry, парсинг ответа\n│   └── storage.py       # SQLite, история, failed audio\n├── templates/\n│   └── index.html       # весь UI — vanilla JS, без сборки\n├── assets/              # иконки\n├── tests/               # 81 тест\n└── .github/workflows/\n    ├── ci.yml           # lint + typecheck + test\n    └── release.yml      # build .exe → GitHub Release\n```\n\n**Стек:** FastAPI · pywebview · pystray · aiosqlite · httpx · Jinja2 · uv · PyInstaller\n\n**Поток транскрибации:**\n\n```\nмикрофон → MediaRecorder (WebM/Opus)\n         → POST /transcribe\n         → base64 + input_audio → OpenRouter API\n         → парсинг TITLE / TEXT\n         → SQLite + возврат в UI\n```\n\nПри ошибке API аудио сохраняется как `FAILED_*.ogg` — доступно для скачивания из UI.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpyramidheadshark%2Fmilvm_stt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpyramidheadshark%2Fmilvm_stt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpyramidheadshark%2Fmilvm_stt/lists"}