10 Commits
Author SHA1 Message Date
sudoeren e58f924579 Match the no-em-dash rule in the comments added here 2026-08-27 21:40:09 +03:00
sudoeren 90ae1690ab Fix the cleanup prompt language and harden detection parsing
Review found the detected language was only switching the glossary rule, not
the base prompt: a detected Turkish recording with an English interface got the
English cleanup prompt with a Turkish glossary footnote. Both now follow the
spoken language, and the detected-language value is guarded against a server
that returns something other than a string. The record reply is covered by a
CLI test.
2026-08-27 21:40:09 +03:00
sudoeren 245f00125e Report the effective speech language on the finished run 2026-08-27 21:40:09 +03:00
sudoeren c5a7fb2410 Document that the speech language is detected by default 2026-08-27 21:40:09 +03:00
sudoeren 7da871c567 Make auto the default speech language and carry what was detected through
New installs start detecting instead of being locked to one language; a stored
value from before this default still wins. The dictation chain asks
transcribe_detected() in auto mode, records the detected code in history as
speech_language, hands it to the cleanup prompt (a detected Turkish recording
gets the Turkish prompt and glossary rule), and reports it on the socket reply.
The stale comment claiming whisper.cpp's -l auto does not detect is corrected.
2026-08-27 21:40:09 +03:00
sudoeren 1bb5c9ebbc Ask whisper.cpp for the detected language when auto mode runs it
The spoken language is only knowable when the transcription model reports it,
and only whisper.cpp does: the hosted endpoints accept auto but never say what
they heard. transcribe_detected() asks for detection exactly where it can be
answered, the local server in auto mode, by requesting verbose_json with
no_language_probabilities switched back on for that one request (the server
runs with -nlp, which keeps the sweep off everything else).
2026-08-27 21:40:09 +03:00
sudoeren 52880bd81a Document the OpenCode Go provider 2026-08-25 21:35:04 +03:00
sudoeren e6b3fcf48f Reach OpenCode Go from the command line
test-key accepts opencode and reports the model count its /models
endpoint offers, ask --provider accepts it and maps --model to its own
setting, doctor reports its key for cleanup on it, and config list
masks the key like the other two.
2026-08-25 21:05:23 +03:00
sudoeren 3c336e7178 Offer OpenCode Go in the settings window
A key row under Keys, a model box in the cleanup tab and a box in the
agent tab, each with its own model list of the models OpenCode Go
serves over /chat/completions. Fetch model list reads whichever cleanup
provider is on screen, and the Turkish strings cover the new rows.
2026-08-25 21:05:21 +03:00
sudoeren 0a6c6128a2 Add OpenCode Go as a cleanup and agent provider
OpenCode Go serves open coding models from an OpenAI-compatible endpoint. It cannot transcribe, so it joins
the two LLM jobs: transcript cleanup and the plain-chat agent. The key
falls back to OPENCODE_API_KEY, and api.chat() stops hardcoding the
OpenRouter name so the same call serves both.
2026-08-25 21:05:18 +03:00