Back to corpus
research noteexperiment writeup candidatescore 18
Phase 3 – Motion, Voice & Phrase Intelligence (Beta)
**Timeline:** Weeks 13-18 (6 weeks) **Status:** Planning **Goal:** Beta release with motion/voice control, phrase recommendations, and UI deck lanes
Full HTML reader
Read the full artifact
Extracted abstract or opening context
**Timeline:** Weeks 13-18 (6 weeks) **Status:** Planning **Goal:** Beta release with motion/voice control, phrase recommendations, and UI deck lanes
Phase 3 transforms Echelon from a core audio engine into a complete performance instrument by adding: - **Motion Control**: DELL motion stream integration for gesture-based deck control - **Voice Control**: Whisper-rs integration for voice commands - **Phrase Intelligence**: Online phrase recommendation service with <5 ms latency - **User Interface**: Deck lanes, phrase browser, automation editor, MIDI learn
This phase enables the core "motion- and voice-driven performance instrument" vision from the project plan.
1. Motion bridge connecting Episode 1 motion pipeline to Echelon scheduler 2. Voice recognizer with command parsing and feedback 3. Phrase intelligence service with FAISS-based recommendations 4. Complete UI with deck lanes, phrase browser, and automation editor 5. End-to-end integration demonstrating motion → phrase → audio flow
- Motion → action latency: <50 ms - Voice recognition accuracy: >90% - Phrase recommendation latency: <5 ms - UI frame rate: 60 FPS - Beta demo: 5-minute performance demonstration
Promotion decision
What has to happen next
Attach run IDs, datasets, metrics, and reproduction commands.
Why this is not always a full paper yet
Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.