feat(tts): add Gemini audio tag rewrite

This commit is contained in:
Barron Roth
2026-06-10 02:57:39 -07:00
committed by Teknium
parent 5718811de0
commit 2c19208224
7 changed files with 297 additions and 20 deletions
+12
View File
@@ -1290,6 +1290,14 @@ DEFAULT_CONFIG = {
"timeout": 30,
"extra_body": {},
},
"tts_audio_tags": {
"provider": "auto",
"model": "",
"base_url": "",
"api_key": "",
"timeout": 30,
"extra_body": {},
},
# Triage specifier — flesh out a rough one-liner in the Kanban
# Triage column into a concrete spec, then promote it to ``todo``.
# Invoked by ``hermes kanban specify`` (single id or --all). Set a
@@ -1575,6 +1583,10 @@ DEFAULT_CONFIG = {
"gemini": {
"model": "gemini-2.5-flash-preview-tts",
"voice": "Kore",
# When true, Gemini 3.1 TTS uses a hidden auxiliary-model rewrite
# pass to insert freeform square-bracket audio tags into the TTS
# script. Visible chat replies are unchanged.
"audio_tags": False,
# Optional local Markdown/text file with Gemini TTS performance
# direction. It may include AUDIO PROFILE, SCENE, DIRECTOR'S NOTES,
# SAMPLE CONTEXT, and either a `{transcript}` placeholder or no