Idea still being shaped
Public Speaking, Multimodally
What kind of rehearsal feedback helps a speaker make a better human choice?
A learner-owned speaking goal guides private rehearsal across one or more modes. Evidence-based observations support a chosen revision without reducing a person to an accent score or personality inference.
Why this idea may matter
Why this might be worth exploring
It centers the speaker’s judgment about useful feedback and keeps recordings under the speaker’s control.
Question to choice
One possible path
- 01
Set a goal
Set a learner-owned goal.
- 02
Rehearse
Rehearse with one or more modes.
- 03
Observe
Collect evidence-based observations.
- 04
Choose a revision
Choose one revision.
- 05
Rehearse again
Rehearse again and decide what feedback to keep.
Possible participant artifact
What might leave the table
A revised short talk and a private evidence-based rehearsal record.
Keep the checking visible
Evidence and checking
- Timestamped observable behavior tied to the learner’s goal, not inferred personality or emotion.
Where AI might help
Help organize, transcribe, compare takes, or surface observable patterns.
What people still decide
Grade accents, infer personality or emotion, assign one score, or replace the speaker’s judgment.
Trust boundaries
- Recordings stay private, exportable, and deletable.
- Synthetic readings are labeled and avoid recognizable-person imitation.
Still a working idea
What is still open
- First speech prompt.
- Which modes appear in the gathering.
- Recording retention policy.
Curiosity Capability Atlas