Add Voicebox-inspired features: effects, history, chunked TTS, playlist, persona, STT refinement
- Audio effects panel (reverb, compressor, chorus, pitch shift via Spotify pedalboard)
with presets (Studio, Broadcast, Telephone, Warm, Radio) — /api/audio/effects endpoint
- Generation history: last 20 sessions, play/reuse/add to playlist
- Chunked TTS: sentence-boundary splitting for unlimited-length text + WAV merge
- Playlist: ordered clip queue with up/down reorder + Export merged WAV
- LLM STT refinement: clean transcripts with local LLM (remove fillers/repetitions/corrections)
— /api/refine-text endpoint, configurable toggles, restore-original button
- Voice persona: character description stored in voice meta.json, shown in inspector,
"Rewrite with persona" button in TryOut via /api/rewrite-with-persona endpoint
- Voices export/import: ZIP backup+restore via /api/voices/export + /api/voices/import
- More whitespace, cleaner UI throughout all new panels
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
mARTin-B78
21a219f816