Appearance
Changelog
Version history and release notes for VoiceOtaku.
v1.0.0
2026-04-24 — Initial release
The first public release of VoiceOtaku — a voice-driven AI anime recommendation assistant.
Features
Voice Call Interface
- One-tap call flow from a single screen with no account or registration required
- Browser-based microphone recording sends audio to the AI pipeline in real time
- AI response played back as synthesised speech and displayed as transcript simultaneously
- Audio waveform visualiser shows microphone activity during a call
AI Recommendation Engine
- Conversational AI understands natural language requests — genres, moods, titles, themes
- Supports follow-up questions within a single session
- Detects conversation end signals (
RECOMMENDATION_COMPLETE,GOODBYE) and closes the call gracefully
Queue System
- Single-slot FIFO queue ensures one caller at a time for consistent quality
- Queue position and estimated wait time displayed in real time, auto-refreshed every few seconds
- JWT-authenticated session tokens scoped to each queue slot
- Processing slot released automatically after the call ends or after the 2-minute timeout — abandoned tabs never stall the queue
Call Timer
- Per-call 2-minute hard cap with a visible countdown timer
- Warning shown at 20 seconds remaining so callers can wrap up naturally
- Session ends automatically when the timer expires
Microphone Selector
- Lists all available audio input devices recognised by the browser
- Device selection persists for the duration of the session
Transcript Display
- User speech and AI responses appear on screen as text alongside the voice output
- Transcript clears between sessions — nothing is stored after the call ends
End Screen
- Brief summary screen shown after each call before returning to the main screen
Deployment
- Single
docker compose up --buildcommand builds and serves the frontend on port 8080 - Build-time environment variables configure the N8N backend URL, call duration, queue polling interval, and request timeout
Version Format
VoiceOtaku follows Semantic Versioning:
| Segment | Meaning |
|---|---|
| MAJOR | Breaking changes to the API or user flow |
| MINOR | New features, backwards-compatible |
| PATCH | Bug fixes and minor improvements |
Upcoming
See the FAQ for a note on planned improvements, or reach out to the team running your instance with feature requests.