Fish Audio Drama 3: direct speech in plain language
Fish Audio Drama 3 is a text-to-speech model built for voice direction. Use square brackets to describe tone, pacing and character, or paired tags to direct particular words and phrases. Results vary with the script and voice.
Choose a compatible Fish Audio voice on the text-to-speech page, then look for Drama 3 in the model list. If it is absent, access is not enabled on the current site. Check the displayed credit quote before generating. Timestamp alignment, experimental language overrides and word-level repair of existing audio are outside this site’s supported scope.
How to direct a performance
Describe the performance
Use ordinary language to describe tone, pacing and character without first choosing fixed audio tags.
Shift mid-sentence
Use paired tags such as <whisper>…</whisper> to direct specific words within a sentence.
Stage multiple characters
Arrange dialogue for different characters within one scene.
Guide pronunciation
Use pronunciation controls for names and terms, then generate the audio again after editing the text.
Where this could help
Direct character dialogue, podcasts and narrated stories with performance instructions. The site integration covers standard generation, long-form content and multi-speaker dialogue with compatible Fish Audio voices.
Can I try it here?
Choose a compatible Fish Audio voice on the text-to-speech page, then look for Drama 3 in the model list. If it is absent, access is not enabled on the current site. Check the displayed credit quote before generating. Timestamp alignment, experimental language overrides and word-level repair of existing audio are outside this site’s supported scope.