Skip to main content

Todos

High Priority​

  • Research streaming ASR solutions (Whisper, Deepgram, etc.) @M1
  • Design turn-taking model architecture @M1
  • Prototype incremental response generation @M1
  • Implement basic conversation loop with streaming input @M1

Medium Priority​

  • Implement interruption detection and handling @M2
  • Build context threading for tracking conversation state @M2
  • Add audio ducking for simultaneous speech @M2
  • Enable bidirectional listening (hear while speaking) @M2
  • Design shared context window interface @M3
  • Implement image drag-and-drop support @M3

Low Priority​

  • Add code file support for shared context @M3
  • Implement AI visual editing capabilities @M3
  • Build split view for multiple files @M3
  • Add document sharing support @M3

Completed​

(No tasks completed yet - project in concept phase)

Ideas / Backlog​

  • Investigate backchanneling responses ("uh-huh", "I see")
  • Research emotion detection from voice
  • Consider multi-participant conversation support
  • Explore integration with Ramble for ASR component
  • Explore integration with Artifact Editor for shared context

Known Issues​

IssueStatusWorkaround
No known issues yetN/AN/A