Rhasspy
Fully offline voice assistant for Home Assistant — runs on a Raspberry Pi with no cloud.
Drop your audio. Transcript in seconds. First transcript free, then $4 = 300 min
Upstream is archived. Rhasspy’s repository has been set read-only by its owner, so there are no further releases, fixes, or security patches. The last commit landed 2023-12-26. It can still be cloned and run, and forks may be maintained elsewhere. See the repository.
Best for home Assistant users who want voice control with zero cloud traffic and full control over wake-word and ASR. Pricing: free.
What it is
Rhasspy is the canonical offline voice assistant for Home Assistant. It chains wake word (Porcupine, openWakeWord, Snowboy) → ASR (Kaldi / Vosk / Whisper) → intent recognition → Home Assistant action. Now maintained by Nabu Casa.
Watch out for: Setup is YAML-heavy; non-trivial to tune for a non-developer; voice quality depends on chosen backend (Kaldi, Pocketsphinx, Vosk, Whisper).
Install / use
rhasspy.readthedocs.io — Docker image or HA Add-on
Features
| Speaker diarization | No |
| Word-level timestamps | No |
| Streaming / real-time | Yes |
| Languages supported | 27 |
| HIPAA eligible | No |
Links
Rhasspy vs Whipscribe
| Feature | Rhasspy | Whipscribe |
|---|---|---|
| Category | Open source | Transcription APIs |
| Pricing | free | $4–$24 one-time packs (credits never expire) · $2 single unlock · 60 min free on signup |
| Speaker diarization | — | Yes |
| Word timestamps | — | Yes |
| Streaming | Yes | No |
| Languages | 27 | 99 |
| Platforms | Linux, Docker, Hardware | Web, API, MCP |
Alternatives to Rhasspy
Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.