Model（Create Voice）

POST

/fish-audio/model

Sound cloning from Fish Audio: submit audio files for cloning.

Requirements for incoming audio:
Audio Length: 30-45 seconds of high-quality audio
Speaker: Single speaker only.
Other details: Consistent volume, pitch, and emotional expression and Short pauses (approximately 0.5 seconds recommended)

Ideal state：
no background noise
professional recording quality
no room echo

Fish Audio：https://docs.fish.audio/api-reference/endpoint/model/create-model

Price：0 PTC/call

Request

Header Params

Authorization

string

optional

Example:

Bearer {{YOUR_API_KEY}}

Body Params multipart/form-data

visibility

string

optional

Available options: public, unlist, private

Example:

private

type

string

required

Model type, tts is for text to speech

Example:

tts

title

string

required

Model title or name

description

string

optional

Model description

cover_image

file

optional

Model cover image, this is required if the model is public

train_mode

string

required

Available options: fast

Example:

fast

voices

file

required

Upload voices files that will be used to tune the model

texts

string

optional

Texts corresponding to the voices, if unspecified, ASR will be performed on the voices

Request samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.302.ai/fish-audio/model' \
--header 'Authorization: Bearer sk-jls4AaVBGoe1GwZD64qZA1qyKTN1MPHa4NmvH1cT68z7K1Zz' \
--form 'visibility="private"' \
--form 'type="tts"' \
--form 'title=""' \
--form 'description=""' \
--form 'cover_image=@""' \
--form 'train_mode="fast"' \
--form 'voices=@""' \
--form 'texts=""' \
--form 'tags=""' \
--form 'enhance_audio_quality="false"'

Responses

🟢200成功

application/json

Body

object {0}

Example

{}

Modified at 2024-09-20 10:19:58

TTS（Text to Speech）

Model（Obtain Voice）