Pi opensources pi-voice: fully local voice pipeline for coding agents
solyarisoftware · x · 2026-09-24
Earendil Works open-sourced pi-voice, a voice plugin for the Pi coding agent, supporting local STT models including Whisper, Qwen3-ASR, Parakeet, Voxtral and SenseVoice, with speech transcribed directly into Pi.
Key use cases:
- Fully local chain: pair Pi with a local model plus pi-voice local STT, and the whole loop from speech to model to code actions skips third-party services—nice for privacy-sensitive setups.
- Voice over typing: dictate ad-hoc instructions while the agent runs in the background (e.g. "check that error, only fix the backend, run tests and report back").
- Audio/video input: transcribefile turns meeting recordings, interviews or tutorial videos into text for Pi to summarize or extract tasks from.
A /voice command is reserved, hinting at a future full voice mode. Pi v0.87.1 also added default support for Claude Opus 5.5, GPT-6 Sol/Luna, and made Grok 4.7 xAI's default.
More from coding & agent
- Notion Engineer: Subsidized Tokens Mean You Should Switch Agents Freely Without Losing Context — nbaschez · 2026-09-24
- Claude Code drops planning mode: spend 95% of your time verifying output, not planning — casper_hansen_ · 2026-09-24
- Unverified claim: 'GPT-6 Astra' builds full video projects via Codex + Dreamina CLI — JaynitMakwana · 2026-09-24
- Open-Source Rust/CUDA Engine reflex Adds DeepSeek MLA, Cold-Start 1.3x Faster Than llama.cpp — SaveAmerica2024 · 2026-09-24
- Building evals for a domain-tuned 4B local LLM: JSON schema plus exact, partial, and semantic matching — funJS · 2026-09-24
- Nerdearla Talk: Building Custom Agents With ADK, MCP and Memory Bank — leslysandra · 2026-09-24