Skip to content

Drag-and-drop files into empty note: transcribe audio or parse transcript #4412

@ComputelessComputer

Description

@ComputelessComputer

Summary

Dragging and dropping a file into an empty note should automatically handle it based on file type:

  • Audio files → transcribe the audio and populate the note
  • Text-based transcript files (e.g. .vtt, .txt, .srt, etc.) → parse the transcript and populate the note

Details

Text-based transcript files can come in many formats with varying structures. We may need to use an LLM to analyze the structure first before parsing, rather than relying on rigid format-specific parsers.

Acceptance criteria

  • Audio files dropped into an empty note trigger transcription
  • Text files dropped into an empty note are parsed into a structured transcript
  • LLM-based structure analysis for ambiguous or unknown text formats
  • Graceful fallback/error handling for unsupported file types

Related

This is related to #2884 and #2854

Metadata

Metadata

Assignees

Projects

Status

Backlog

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions