feat: add first-party voice transcription with local Whisper

[cmdNiels]

What does this PR do?

Implements first-party, local voice transcription using Whisper.cpp via @huggingface/transformers. All processing happens locally with zero configuration required.

Features

• Privacy-first: All transcription happens locally, no external API calls
• Zero configuration: Works out of the box after enabling in /status
• Offline support: Works completely offline after initial model download
• Three model sizes: tiny (75MB), base (142MB, default), small (466MB)
• Customizable keybind: Default \ for recording, configurable in ~/.config/opencode/opencode.json

This follows the frontend-backend separation design of OpenCode:

• TUI: Uses sox/ffmpeg for microphone recording (tested on Linux)

Usage

1. Enable voice in /status dialog
2. Wait for model download (one-time, ~142MB for base model)
3. Press \ to start recording
4. Speak clearly
5. Press \ again to stop
6. Transcribed text appears in prompt input

Models are cached in ~/.cache/opencode/models/ and persist across sessions.

Comparison with PR #9264

PR #9264 implements voice via external APIs (Groq/OpenAI/local whisper-fastapi). This implementation prioritizes privacy and simplicity. Both approaches have merit. This implementation aligns with OpenCode's philosophy of local-first, privacy-respecting tooling.

Fixes#2425
Fixes#4695

How did you verify your code works?

• All 792 tests pass, TypeScript compiles cleanly
• Manually tested: recording with \ keybind, model download/caching, transcription accuracy
• Verified config persistence, auto-start, offline mode with cached models
• Tested on Linux with sox/ffmpeg (macOS/Windows unverified but should work)

[github-actions]

The following comment was made by an LLM, it may be inaccurate:

Potential Related PRs Found

1. PR #9264 - feat: voice-typing using external Whisper / ALM API

• feat: voice-typing using external Whisper / ALM API #9264
• Relationship: This is explicitly mentioned in the PR description as a prior approach. While PR feat: add first-party voice transcription with local Whisper #11345 uses local Whisper.cpp with privacy-first design, PR feat: voice-typing using external Whisper / ALM API #9264 implements voice via external APIs (Groq/OpenAI/local whisper-fastapi). Both address the same feature request but with different implementation philosophies.

2. PR #3827 - Add voice-to-text transcription feature

• Add voice-to-text transcription feature #3827
• Relationship: Earlier attempt at voice-to-text functionality that may be related to the same feature request. Worth checking if it's been closed/superseded by either of the newer PRs.

Note: PR #11345 (the current PR) explicitly references PR #9264 in its description and provides a comparison, so the relationship is acknowledged by the authors.

[shenron0101]

Eagerly waiting for this PR to be merged!

[kungfusaini]

Also can't wait for this!!

[MartinWie]

any update here ?

[JorySeverijnse]

Can we at least have an update as to why this is not being merged after more than a month?

[hajnalmt]

CC @rekram1-node@thdxr The original issues and PRs were assigned to you, this PR seems vastly superior to those. It implements first-party voice transcription and appears to address the original feature requests

[Galtozzy]

CC @thdxr

Any updates on this one?

[SimSonic]

Any news?

[kuroneko83]

Still nothing?

[AidenGeunGeun]

rip