fix: TTS ack-bank stale voice cache + website/English pronunciation #6

2026-08-30T06:39:25Z

arystanbek commented

2026-08-30 06:39:25 +00:00

RuntimeConfiguredTTSProvider never overrode cache_fingerprint(), так что prebaked ack-фразы ("Секунду" и т.п.) кэшировались без учёта смены голоса и продолжали звучать старым голосом ElevenLabs после смены voice id, пока обычные реплики LLM уже звучали новым. Плюс инструкция для voice-модели не оставлять сайты/английские слова латиницей (egov.kz читалось как эговкз).

arystanbek added 1 commit 2026-08-30 06:39:26 +00:00

fix: TTS ack-bank served stale voice after voice-config changes; add website/English pronunciation rules fe8598e09f

RuntimeConfiguredTTSProvider (the live 'dynamic' TTS provider, resolves
voice from DB-backed config) never overrode cache_fingerprint(), so it
fell back to the base class's empty string. PrebakedAckBank keys its
on-disk cache on that fingerprint, so short filler phrases like
'Секунду' kept serving audio baked with the previous ElevenLabs voice
even after a voice change, while full LLM replies (cached inside the
resolved provider itself, keyed on its own voice id) already used the
new voice — explaining why callers heard two different voices in the
same call. Fix: delegate cache_fingerprint() to the resolved provider.

Also extend the voice delivery_hint so the model transliterates website
addresses and English words/abbreviations into spoken Cyrillic instead
of leaving raw Latin text for the TTS engine to mangle (egov.kz was
coming out as 'эговкз').

arystanbek merged commit 0a829d26c7 into main

2026-08-30 06:39:33 +00:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: digiops/call-center#6