10 open-source repos for offline speech: Whisper, Argos, Piper, Vosk and more
Shruti_0810 · x · 2026-10-22
A roundup of open-source repos for building offline language AI: Whisper for transcription, Seamless Communication for speech translation, Argos Translate and Bergamot for offline/private text translation, Piper and Coqui TTS for local speech synthesis, and Vosk for offline speech recognition. The attached image highlights F5-TTS, a flow-matching TTS project with 15k stars.
More from Apps
- Recursive releases Flint, points to official update page — lazyniu · 2026-10-22
- Gemini image bug: every download returns the first generated image, users fume — WaajibUlCuddle · 2026-10-22
- Resend Launches Inboxes, Purpose-Built Mailboxes for AI Agents — philnash · 2026-10-22
- Xaira Unveils X-Scientist: AI Nominates Drug Targets ~10x Faster Than Humans — BoWang87 · 2026-10-22
- Building AI as a study partner, not a philosopher simulator: labeling supported vs extrapolated claims — 23Windy1City33 · 2026-10-22
- Solo dev builds 2,000-hero roguelite in 4 months with Claude Code running the whole pipeline — R3dNaX7 · 2026-10-22