voice on Hacker News

98 titles from 83 authors in the last 30 days, with 94 distinct links and 27 Show HN.

Counts titles, it does not confirm facts. Open the source before concluding anything.

Projects on this term

The score comes from our model.

• Show HN: Dub Anything – Dub videos in each actor's own voice · score 7.8

  Dub Anything uses an AI coding agent to direct high-quality video dubbing line by line, preserving each actor's own voice, delivery timing, and original background audio. It stands out for its agent-directed approach, per-line voice cloning prompted by the actor's own delivery, and automated QA that re-transcribes every rendered line to catch errors.

• Show HN: Jarvis, an Android voice assistant that drives other apps by itself · score 7.8

  JARVIS is an experimental Android voice assistant that can drive other apps via an AccessibilityService operator, control smart TVs and PCs, and manage memories and routines. It integrates an on-device TypeScript brain with Tauri, React, and Kotlin, supporting multiple LLM providers and innovative features like smart model routing and wake-word detection. The project is ambitious and technically complex, though still rough around the edges.

• Show HN: Whospair – a serverless Android messenger over Tor, with voice calls · score 7.5

  Whospair is a serverless Android messenger that uses Tor hidden services for direct peer-to-peer communication, eliminating central servers and accounts. It combines strong end-to-end encryption, voice calls, and advanced privacy features like duress PIN and face-detection screen protection, all packaged in a single APK with self-updating over Tor.

• Show HN: Macvoice – A local voice agent for Mac in 1.4 GB, no PyTorch · score 7.3

  Macvoice is a low-latency local voice agent pipeline for Apple Silicon that runs in under 2 GB without PyTorch, combining Silero VAD, Smart Turn, Parakeet ASR, an LLM, and Phonon TTS. It is notable for its careful engineering: echo cancellation via AVAudioEngine, streaming resampling, unit-tested endpointing, and a fully offline pipeline with benchmark comparisons.

Recent titles

1. We build our voice input method like an experiment
   2026-11-01 · arion.astrafoundry.si · HN discussion
2. Show HN: Kin – Natural language search for family voice notes and memories
   2026-11-01 · kin.camera · HN discussion
3. Hex – local voice dictation for Mac
   2026-10-31 · anoma.ly · HN discussion
4. Show HN: Hey Spark – open-source voice assistant for your laptop, any LLM brain
   2026-10-31 · github.com · HN discussion
5. Show HN: I benchmarked a voice AI against real people. It scored at chance
   2026-10-31 · mundi.chat · HN discussion
6. Measuring voice agents: turn-taking, tool calls and expressive speech
   2026-10-31 · soniqo.audio · HN discussion
7. Show HN: Deckdrill – Pimsleur-style audio drills from Anki decks, in your voice
   2026-10-30 · deckdrill.app · HN discussion
8. Personal Agent's Voice Security
   2026-10-30 · github.com · HN discussion
9. Show HN: Anna – a voice assistant that reports your machine's load, RAM and disk
   2026-10-29 · github.com · HN discussion
10. CPU-only voice clone that transcribes each sentence and retakes misreads
    2026-10-29 · blakelapierre.substack.com · HN discussion
11. LLM Voice
    2026-10-29 · martinfowler.com · HN discussion
12. Show HN: Macvoice – A local voice agent for Mac in 1.4 GB, no PyTorch
    2026-10-29 · github.com · HN discussion
13. TTS-1 from AssemblyAI – Frontier voice at $15/M chars
    2026-10-28 · assemblyai.com · HN discussion
14. Optimizing voice activity detection hyperparameters
    2026-10-28 · blog.ricky0123.com · HN discussion
15. I combined codex live voice with mobile use and the result is amazing
    2026-10-28 · github.com · HN discussion
16. Comodity Software: Voice Transcription
    2026-10-28 · dejang.github.io · HN discussion
17. Tenstorrent and Smallest.ai Partner to Bring Production-Grade Voice AI On-Prem
    2026-10-28 · tenstorrent.com · HN discussion
18. Dhwani: A self-hosted voice notes app built with Whistle by Cactus Compute
    2026-10-28 · github.com · HN discussion
19. Block's new voice model sounds eerily human
    2026-10-28 · techcrunch.com · HN discussion
20. Turn 40: benchmarking voice agents over a long session
    2026-10-27 · presence.bohita.com · HN discussion
21. Can Machines Follow One Voice Through a Crowded Room?
    2026-10-27 · kelvinjiang.com · HN discussion
22. Voxtral TTS vs. ElevenLabs: Voice Cloning and TTS
    2026-10-27 · techstackups.com · HN discussion
23. Twins AI moved its production AI voice platform from GCP to Azure
    2026-10-26 · twinsai.com · HN discussion
24. Eric Bauza, Voice of Bugs Bunny and Daffy Duck, Dies at 46
    2026-10-26 · washingtonpost.com · HN discussion
25. Partyline: A group voice application for Reticulum, built on LXST
    2026-10-26 · github.com · HN discussion
26. How to take back your voice
    2026-10-24 · newsletter.openhomefoundation.org · HN discussion
27. Ramblr (FOSS) – Tap-to-dictate voice input for Android
    2026-10-24 · github.com · HN discussion
28. Show HN: ZapFast, a native WhatsApp client with voice and video calls on Linux
    2026-10-23 · zapfast.rocks · HN discussion
29. How to Use ZizkaDB with LiveKit to Audit Your Voice Agents
    2026-10-23 · medium.com · HN discussion
30. Show HN: Moin – A P2P voice app built on iroh and Media over QUIC
    2026-10-22 · moin.chat · HN discussion

All trends