{"id":48801168,"url":"https://github.com/sophiie-ai/whispur","last_synced_at":"2026-04-14T03:01:05.449Z","repository":{"id":350937379,"uuid":"1205540118","full_name":"sophiie-ai/whispur","owner":"sophiie-ai","description":"Menu-bar voice dictation for macOS with local-first Apple STT, multi-provider speech-to-text, and optional AI cleanup.","archived":false,"fork":false,"pushed_at":"2026-04-12T21:04:28.000Z","size":2131,"stargazers_count":0,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-12T22:24:33.950Z","etag":null,"topics":["ai","dictation","macos","menu-bar","speech-to-text","swift","voice"],"latest_commit_sha":null,"homepage":null,"language":"Swift","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/sophiie-ai.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","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":"2026-04-09T03:52:22.000Z","updated_at":"2026-04-12T21:04:20.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/sophiie-ai/whispur","commit_stats":null,"previous_names":["sophiie-ai/whispur"],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/sophiie-ai/whispur","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sophiie-ai%2Fwhispur","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sophiie-ai%2Fwhispur/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sophiie-ai%2Fwhispur/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sophiie-ai%2Fwhispur/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sophiie-ai","download_url":"https://codeload.github.com/sophiie-ai/whispur/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sophiie-ai%2Fwhispur/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31779947,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-14T02:24:21.117Z","status":"ssl_error","status_checked_at":"2026-04-14T02:24:20.627Z","response_time":153,"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":["ai","dictation","macos","menu-bar","speech-to-text","swift","voice"],"created_at":"2026-04-14T03:01:03.915Z","updated_at":"2026-04-14T03:01:05.439Z","avatar_url":"https://github.com/sophiie-ai.png","language":"Swift","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Whispur\n\nWhispur is a macOS menu-bar dictation app that turns speech into polished text and drops it straight back into the app you are already using.\n\nHold a shortcut to talk, or use an optional toggle shortcut for hands-free capture. Whispur records, transcribes, optionally cleans up the transcript with an LLM, and pastes the result where your cursor is.\n\n## Screenshots\n\nScreenshots will live in [`docs/screenshots`](docs/screenshots/). A placeholder capture script is available at [`scripts/generate-screenshots.sh`](scripts/generate-screenshots.sh) until the final product shots are added.\n\n## Features\n\n- Lives in the macOS menu bar instead of taking over your desktop\n- Hold-to-talk and toggle dictation modes\n- Multi-provider speech-to-text with local Apple dictation support\n- Optional transcript cleanup with provider-selectable LLMs\n- Paste-back into the active app with clipboard preservation\n- Custom vocabulary and cleanup prompts for better technical dictation\n- Local-first default path when you stick with Apple on-device transcription\n- Sparkle-based auto-updates for signed releases\n\n## Install\n\n### DMG download\n\nDownload the latest signed DMG from [GitHub Releases](https://github.com/sophiie-ai/whispur/releases).\n\n### Homebrew cask\n\nHomebrew installation is planned but not published yet.\n\n### Build from source\n\n```bash\ngit clone https://github.com/sophiie-ai/whispur.git\ncd whispur\nbrew install xcodegen create-dmg\nmake all\nmake run\n```\n\n## Setup\n\n1. Launch Whispur from Applications. It opens as a menu-bar app.\n2. Open `Settings` from the menu bar.\n3. Complete the setup checklist for microphone and Accessibility access.\n4. Choose your speech provider.\n5. Add API keys for any cloud providers you want to use.\n6. Optionally add an LLM provider for transcript cleanup.\n7. Review your hold and toggle shortcuts.\n8. Dictate into any focused text field.\n\n## Usage\n\n- Hold shortcut: Press and hold your configured shortcut to record, then release to transcribe and paste.\n- Toggle shortcut: Press once to start recording, press again to stop and process.\n- Activity log: Open `Settings` → `Activity` to inspect raw and cleaned transcripts.\n- Prompt tuning: Open `Settings` → `Prompts` to change cleanup behavior or add domain vocabulary.\n\n## Provider Matrix\n\n### Speech-to-text\n\n| Provider | Status | Notes |\n| --- | --- | --- |\n| Apple Speech Recognition | Available | On-device and local |\n| OpenAI Whisper | Available | Cloud STT via API key |\n| Deepgram | Available | Cloud STT via API key |\n| ElevenLabs Scribe | Available | Cloud STT via API key |\n| AWS Bedrock | Planned | Provider slot exists, implementation not shipped yet |\n\n### Transcript cleanup\n\n| Provider | Status | Notes |\n| --- | --- | --- |\n| Anthropic Claude | Available | High-quality cleanup |\n| OpenAI | Available | Fast cleanup path |\n| Groq | Available | OpenAI-compatible endpoint |\n| AWS Bedrock | Planned | Provider slot exists, implementation not shipped yet |\n\n## Privacy\n\nAudio stays local by default when you use the built-in Apple speech provider. If you switch to a cloud STT or LLM provider, the audio or transcript needed for that provider is sent only to the services you configure.\n\nWhispur stores API keys in the macOS Keychain.\n\n## Development\n\n```bash\nmake generate\nmake all\nmake dmg\n```\n\nKey project areas:\n\n- `Sources/App`: app lifecycle, menu-bar scenes, state\n- `Sources/Audio`: recording and normalization\n- `Sources/Providers`: STT and LLM integrations\n- `Sources/Pipeline`: dictation orchestration\n- `Sources/Input`: shortcuts and paste-back\n- `Sources/UI`: menu bar, onboarding, settings, about window\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md).\n\n## License\n\nWhispur is released under the [MIT License](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsophiie-ai%2Fwhispur","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsophiie-ai%2Fwhispur","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsophiie-ai%2Fwhispur/lists"}