ElebobaandClaude Opus 5.5 d700608825 fix(llm): STT prompt that keeps unclear speech; reasoning off (~60 s → ~3 s) ...

Live check on synthetic speech (espeak) via OpenRouter gemini-2.5-flash:
3–5 s per clip, silence → no speech.

Refs #116

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-31 11:29:40 +03:00

..

api

feat(llm): speech-to-text for patient voice input via OpenRouter (+ offline mock)

2026-10-31 11:19:55 +03:00

channels

feat(bot): /start without a code on the demo stand opens a private copy of veteran Andrey

2026-10-05 12:28:03 +03:00

core

feat: crisis message after the test and non-blocking; weapon answer triggers nothing; ЕМИАС demo = Никита

2026-10-04 23:30:05 +03:00

db

feat: presentation demo mode — one-click private demo patient, screening on/off switch

2026-10-04 21:01:05 +03:00

dialog

feat: crisis message after the test and non-blocking; weapon answer triggers nothing; ЕМИАС demo = Никита

2026-10-04 23:30:05 +03:00

integrations

fix(bot): Telegram through TG_PROXY_URL — docker on the RF server has no IPv4 route to Telegram

2026-10-04 23:40:53 +03:00

llm

fix(llm): STT prompt that keeps unclear speech; reasoning off (~60 s → ~3 s)

2026-10-31 11:29:40 +03:00

scheduler

fix(clinical): red flags, one crisis rule, callback after a crisis, one table of numbers

2026-10-04 16:21:16 +03:00

services

fix: chat replies in any situation — no false crisis re-fire, answers in crisis, medicine questions anywhere

2026-10-04 18:23:03 +03:00

__init__.py

chore(devops): project skeleton — uv, docker compose, Makefile, CI

2026-10-03 14:46:01 +03:00

config.py

feat(llm): speech-to-text for patient voice input via OpenRouter (+ offline mock)

2026-10-31 11:19:55 +03:00

main.py

feat(contract): add opt-in MAX configuration and lifecycle

2026-10-04 15:58:25 +03:00