# Fish Audio emotion audition sheet

Model name / version:
Voice / generation settings:
Matching syntax: S1 parentheses; S2-family square-bracket descriptions.

Spoken transcript: That's the last box. We can go home.
Intention: tired relief after finishing a long packing session.

| Candidate | Cue | Feeling heard without label | Words clear? | Duration fits? | Decision |
|---|---|---|---|---|---|
| Plain | None | | | | |
| A | [relieved] | | | | |
| B | [tired but relieved] | | | | |

The cues are proposed S2-family experiments, not guaranteed presets or recorded results.

- [ ] Same voice, words, model, and settings for the first comparison.
- [ ] Playback levels are similar.
- [ ] “Last box” remains intelligible.
- [ ] The second sentence sounds like release, not a question unless intended.
- [ ] Any script rewrite starts a new comparison.
- [ ] Performance tags are removed from the spoken transcript used for captions.
- [ ] Final take reviewed under the actual picture and music.

Chosen take / reason:
Reusable direction for this narrator: