diff options
| author | Danilo M. <danix@danix.xyz> | 2026-10-19 11:20:18 +0200 |
|---|---|---|
| committer | Danilo M. <danix@danix.xyz> | 2026-10-19 11:20:18 +0200 |
| commit | 393e2ccc4eb3bb3c279b7d60652a4380e701c065 (patch) | |
| tree | e2bb91a7e4adcb9a06c3834a9e878f88e4684ea7 | |
| parent | 1b3dca845f2b2866cc967e4f5fb5f21a6d866a4e (diff) | |
| download | desktop-assistant-393e2ccc4eb3bb3c279b7d60652a4380e701c065.tar.gz desktop-assistant-393e2ccc4eb3bb3c279b7d60652a4380e701c065.zip | |
docs: voice input works end to end with the canvas mic
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
| -rw-r--r-- | AGENTS.md | 7 |
1 files changed, 4 insertions, 3 deletions
@@ -6,7 +6,10 @@ read-only tools and a SQLite history. Nothing leaves the machine except what an enabled tool fetches. Status (2026-10-16): **text daemon**. The front end is the quickshell -canvas (`desktop/assistant/` in the quickshell repo, SUPER+Home). Spec: +canvas (`desktop/assistant/` in the quickshell repo, SUPER+Home). Voice +input (2026-10-19): the canvas's mic records a WAV, the loaded model +transcribes it, and the words are asked as text; no whisper. Spec: +`docs/superpowers/specs/2026-10-19-voice-input-design.md`. Spec: `docs/superpowers/specs/2026-10-16-text-daemon-design.md`. The voice version is the tag `voice-v1`: push-to-talk on the mic's mute @@ -769,8 +772,6 @@ kokoro-onnx GitHub release `model-files-v1.0`. - **Local memory:** `docs/superpowers/plans/2026-09-30-local-memory.md` still applies; its tasks 1-4 target this daemon, the voice-specific bits of task 4 aside. -- **Voice input:** `docs/superpowers/specs/2026-10-19-voice-input-design.md`, - the model transcribes, no whisper; the mic is the canvas's part. - **llamachat ports**, if the canvas replaces llamachat for good: attachments, skills, a model picker, model-written titles. - **Phase 2 tools:** tools that write or modify (events, messages, |