How do you specify which voice to use in API calls? | Fish Audio | Page 1

fathom bone Oct 14, 2025, 3:38 AM

#

curl -X POST https://api.fish.audio/v1/tts \ -H "Authorization: Bearer a44b36eebda94213ad63a797b3c67e1a" \ -H "Content-Type: application/json" \ -H "model: s1" \ -d '{ "text": "Hello! Welcome to Fish Audio. \ This is my first AI-generated voice.", "format": "mp3" }' \ --output welcome.mp3

sly raft Oct 21, 2025, 8:18 AM

#

Seems to me it's not possible currently... Did you find a solution?

mighty parrot Oct 21, 2025, 11:57 AM

#

for me its exactly the same, i can't find any voice id

spiral marsh Oct 21, 2025, 4:33 PM

#

hi there, voices can be specified with the reference_id option.

https://docs.fish.audio/api-reference/endpoint/openapi-v1/text-to-speech#body-reference-id

the reference id is the part that goes after /m/ on fish audio's website, or you can copy model id from the model page:

https://fish.audio/m/06973b7353dd4dda86b7b46924fa01a6/