{"id":47759613,"url":"https://github.com/jacopone/whisper-dictation","last_synced_at":"2026-04-03T05:11:16.428Z","repository":{"id":338688496,"uuid":"1068082535","full_name":"jacopone/whisper-dictation","owner":"jacopone","description":"🎤 Privacy-first local speech-to-text dictation for NixOS - Whisper.cpp powered push-to-talk with real-time feedback","archived":false,"fork":false,"pushed_at":"2026-02-15T23:35:26.000Z","size":66,"stargazers_count":6,"open_issues_count":0,"forks_count":3,"subscribers_count":1,"default_branch":"master","last_synced_at":"2026-02-16T04:27:37.244Z","etag":null,"topics":["dictation","gnome","nixos","privacy","speech-to-text","voice-input","wayland","whisper"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/jacopone.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","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,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-01T20:36:33.000Z","updated_at":"2026-02-15T23:35:29.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/jacopone/whisper-dictation","commit_stats":null,"previous_names":["jacopone/whisper-dictation"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/jacopone/whisper-dictation","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacopone%2Fwhisper-dictation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacopone%2Fwhisper-dictation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacopone%2Fwhisper-dictation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacopone%2Fwhisper-dictation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jacopone","download_url":"https://codeload.github.com/jacopone/whisper-dictation/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacopone%2Fwhisper-dictation/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31335353,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-03T04:42:29.251Z","status":"ssl_error","status_checked_at":"2026-04-03T04:42:12.667Z","response_time":107,"last_error":"SSL_read: 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":["dictation","gnome","nixos","privacy","speech-to-text","voice-input","wayland","whisper"],"created_at":"2026-04-03T05:11:15.648Z","updated_at":"2026-04-03T05:11:16.422Z","avatar_url":"https://github.com/jacopone.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Whisper Dictation\n\nPrivacy-first local speech-to-text for NixOS -- whisper.cpp powered, push-to-talk, paste anywhere.\n\n## Features\n\n- **100% local and private** -- no cloud, no telemetry, works fully offline\n- **Push-to-talk** -- hold Super+Period, speak, release to paste text\n- **Real-time feedback** -- floating GTK4 window shows transcription status\n- **Multilingual** -- supports 99 languages with auto-detection\n- **Wayland native** -- built for GNOME on Wayland, works in any application\n- **Optimized for technical speech** -- tuned for developer and AI workflows\n\n## Requirements\n\n- NixOS or any Linux distribution with Nix\n- Wayland compositor (GNOME recommended)\n- PulseAudio or PipeWire\n- User must be in the `input` group for keyboard monitoring\n\n## Installation\n\n### NixOS (recommended)\n\nAdd to your `flake.nix`:\n\n```nix\n{\n  inputs.whisper-dictation.url = \"github:jacopone/whisper-dictation\";\n\n  # In your configuration\n  environment.systemPackages = [\n    inputs.whisper-dictation.packages.${system}.default\n  ];\n\n  # Enable auto-start\n  systemd.user.services.whisper-dictation = {\n    enable = true;\n    wantedBy = [ \"graphical-session.target\" ];\n  };\n}\n```\n\n### Manual\n\n```bash\ngit clone https://github.com/jacopone/whisper-dictation.git\ncd whisper-dictation\nnix develop\npython -m whisper_dictation.daemon\n```\n\n**First-time setup:** ensure your user is in the `input` group (`sudo usermod -aG input $USER`, then log out and back in), download a Whisper model to `~/.local/share/whisper-models/`, and start the `ydotoold` daemon. See the [first-time setup section in DEVELOPMENT.md](DEVELOPMENT.md) for details.\n\n## Usage\n\nStart the daemon and dictate:\n\n```bash\nrun-daemon          # use config file settings\nrun-daemon-en       # English only (fastest)\nrun-daemon-it       # Italian only\nrun-daemon-auto     # auto-detect language (adds ~1-2s)\n```\n\nThen in any application:\n\n1. Click in a text field\n2. Hold **Super+Period**\n3. Speak naturally\n4. Release the key -- text is pasted instantly\n\nOverride settings per-session with command-line flags:\n\n```bash\npython -m whisper_dictation.daemon --verbose --language auto --model base\n```\n\n## Configuration\n\nEdit `~/.config/whisper-dictation/config.yaml`. Key settings:\n\n- `whisper.model` -- model size: `tiny`, `base` (recommended), `small`, `medium`, `large`\n- `whisper.language` -- language code (`en`, `it`, `auto`, etc.)\n- `hotkey.key` / `hotkey.modifiers` -- push-to-talk keybinding\n\nSee `config.yaml` in the repository for all available options.\n\n\u003cdetails\u003e\n\u003csummary\u003eModel selection guide\u003c/summary\u003e\n\n| Model  | Size   | Speed    | Accuracy | Use Case                  |\n|--------|--------|----------|----------|---------------------------|\n| tiny   | 39 MB  | ~1-2s    | 60%      | Quick notes, testing      |\n| base   | 142 MB | ~4-6s    | 70%      | Recommended for speed     |\n| small  | 466 MB | ~10-15s  | 80%      | Balanced performance      |\n| medium | 1.5 GB | ~20-30s  | 85%      | High accuracy             |\n| large  | 2.9 GB | ~40-60s  | 90%      | Maximum accuracy          |\n\nTimes measured on CPU (4 threads). GPU acceleration can reduce times by 5-10x.\n\n\u003c/details\u003e\n\n## How It Works\n\n1. **Keyboard monitoring** -- `evdev` captures low-level key events\n2. **Audio recording** -- `ffmpeg` records microphone input while the key is held\n3. **Transcription** -- `whisper.cpp` processes audio locally on your machine\n4. **Text insertion** -- `ydotool` pastes transcribed text into the active window\n5. **UI feedback** -- GTK4 floating window shows real-time status\n\n## Comparison\n\n| Feature           | Whisper Dictation | Aqua Voice  | Talon Voice  |\n|-------------------|-------------------|-------------|--------------|\n| Privacy           | Local             | Cloud       | Local        |\n| Cost              | Free              | $8/mo       | $15/mo       |\n| NixOS support     | Native            | No          | Manual       |\n| Technical terms   | 65-85%            | 97%         | 95%          |\n| Wayland           | Yes               | Limited     | X11 only     |\n| Real-time         | Yes               | Yes         | Yes          |\n\n## Development\n\nSee [DEVELOPMENT.md](DEVELOPMENT.md) for the full development guide.\n\n## Troubleshooting\n\nSee [TROUBLESHOOTING.md](TROUBLESHOOTING.md) for solutions to common issues (audio, keyboard detection, ydotool, hotkeys, performance).\n\n## Contributing\n\nContributions welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.\n\n## License\n\nMIT License -- see [LICENSE](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjacopone%2Fwhisper-dictation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjacopone%2Fwhisper-dictation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjacopone%2Fwhisper-dictation/lists"}