{"id":51337369,"url":"https://github.com/evgeniyarbatov/vibing","last_synced_at":"2026-07-02T04:02:26.736Z","repository":{"id":356449674,"uuid":"1232531877","full_name":"evgeniyarbatov/vibing","owner":"evgeniyarbatov","description":"Local voice-to-clipboard transcription on macOS — tap a key to record, get clean text on your clipboard, no cloud needed.","archived":false,"fork":false,"pushed_at":"2026-06-25T11:43:01.000Z","size":61,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-25T13:19:37.539Z","etag":null,"topics":["clipboard","faster-whisper","local-ai","macos","offline","ollama","privacy","speech-to-text","transcription","voice-recorder"],"latest_commit_sha":null,"homepage":"","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/evgeniyarbatov.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":"ROADMAP.md","authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-08T02:54:47.000Z","updated_at":"2026-06-25T11:43:05.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/evgeniyarbatov/vibing","commit_stats":null,"previous_names":["evgeniyarbatov/vibing"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/evgeniyarbatov/vibing","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/evgeniyarbatov%2Fvibing","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/evgeniyarbatov%2Fvibing/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/evgeniyarbatov%2Fvibing/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/evgeniyarbatov%2Fvibing/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/evgeniyarbatov","download_url":"https://codeload.github.com/evgeniyarbatov/vibing/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/evgeniyarbatov%2Fvibing/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35032145,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-02T02:00:06.368Z","response_time":173,"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":["clipboard","faster-whisper","local-ai","macos","offline","ollama","privacy","speech-to-text","transcription","voice-recorder"],"created_at":"2026-07-02T04:02:26.017Z","updated_at":"2026-07-02T04:02:26.730Z","avatar_url":"https://github.com/evgeniyarbatov.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# vibing\n\nPress the **right Option key** to start recording. Press it again to stop. Your speech is transcribed, cleaned up, and copied to the clipboard — all locally, no cloud needed.\n\nInspired by [Vibing](https://vibingjustspeakit.github.io/Vibing/).\n\n## Pipeline\n\n```\nmic → raw WAV → ffmpeg loudnorm → whisper small → ollama mistral-nemo → clipboard\n```\n\n| Step | Tool | Output |\n|------|------|--------|\n| Record | sounddevice (Python) | `data/raw-audio/TIMESTAMP.wav` |\n| Normalize | ffmpeg `loudnorm` | `data/normalized-audio/TIMESTAMP.wav` |\n| Transcribe | faster-whisper `small` | `data/raw-transcript/TIMESTAMP.txt` |\n| Clean up | ollama `mistral-nemo` | `data/clean-transcript/TIMESTAMP.txt` |\n| Deliver | pbcopy | clipboard |\n\n---\n\n## Prerequisites\n\n### 1. ffmpeg\n\n```bash\nbrew install ffmpeg\n```\n\n### 2. faster-whisper\n\n```bash\npip install faster-whisper\n```\n\nThe first transcription downloads the `small` model (~244 MB) automatically.\n\n### 3. Ollama + mistral-nemo\n\n```bash\n# Install ollama\nbrew install ollama\n\n# Pull the model\nollama pull mistral-nemo\n\n# Start the server\nollama serve\n```\n\n### 4. Python 3.10+\n\n```bash\nbrew install python\n```\n\n---\n\n## Configuration\n\nEdit **`config.json`** in the project root. Changes take effect on the next restart.\n\n```json\n{\n  \"ollama_model\": \"mistral-nemo\",\n  \"ollama_prompt\": \"Clean up this voice transcription. ... {transcription}\",\n  \"output_dir\": \"~/Documents/vibing\"\n}\n```\n\n| Key | Default | Description |\n|-----|---------|-------------|\n| `ollama_model` | `mistral-nemo` | Any model available in your local ollama (`ollama list`) |\n| `ollama_prompt` | *(see file)* | Prompt sent to ollama. Must contain `{transcription}` — that placeholder is replaced with the raw whisper output |\n| `output_dir` | `~/Documents/vibing` | Root for all four output subdirectories. Relative paths are resolved from the project root; absolute paths are used as-is |\n\n## Quick start\n\n```bash\nmake run\n```\n\nRuns the recorder in your terminal. Log output appears in real time and is also written to `logs/recorder.log`. Press **right Option** to start/stop recording. Stop with `Ctrl-C`.\n\n### Grant Accessibility access (required)\n\n`pynput` needs the Accessibility permission to read global keyboard events.\n\n1. Open **System Settings → Privacy \u0026 Security → Accessibility**.\n2. Click **+** and add `.venv/bin/python3` inside this project directory.\n   - If running from Terminal, add **Terminal** (or iTerm2) instead.\n\nWithout this step the key listener silently does nothing.\n\n### Makefile targets\n\n| Command | What it does |\n|---------|-------------|\n| `make setup` | Create virtualenv and install Python dependencies |\n| `make run` | Start the recorder in the foreground |\n| `make test` | Run the test suite |\n| `make clean` | Remove virtualenv and all captured data |\n\n## Usage\n\n| Action | Gesture |\n|--------|---------|\n| Start recording | Tap right Option key once |\n| Stop recording | Tap right Option key again |\n\nThe cleaned transcript is always saved to `data/clean-transcript/` even if you use the clipboard version.\n\n## Troubleshooting\n\n### No key events detected\n→ Accessibility permission not granted. See [Grant Accessibility access](#grant-accessibility-access-required).\n\n### faster-whisper import error\n→ Run `pip install faster-whisper` (or `make setup`) and make sure your virtualenv is active.\n\n### ollama connection refused\n→ Start ollama: `ollama serve` or install it as a service via `brew services start ollama`.\n\n### First transcription is slow\n→ Whisper downloads the `small` model (~244 MB) on first use. Subsequent runs are fast.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fevgeniyarbatov%2Fvibing","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fevgeniyarbatov%2Fvibing","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fevgeniyarbatov%2Fvibing/lists"}