POST Method Request
POST
https://tts.nirvanaworld.cn/tts
Headers
Content-Type
application/json
Authorization
Bearer <token>
Body
text
string
Yes
Text to be processed is recommended to be URL-encoded.
character
string
Yes
Role Folder Name: Pay attention to case sensitivity, full-width/half-width characters, and language.
emotion
string
No
Role Emotion: Needs to be an emotion actually supported by the role; otherwise, the default emotion will be used.
text_language
string
No
Text Language: Chinese, English, Japanese, Chinese-English Mixed, Japanese-English Mixed, Multilingual Mixed. Default is Multilingual Mixed.
format
string
No
Format: Default is WAV. Supports WAV, MP3, FLAC.
stream
bool
No
Streaming: When set to true, audio will be returned sentence by sentence. Default is false.
save_temp
bool
No
Save Temporary File: When set to true, the backend will save the generated audio. Subsequent identical requests will directly return this data without waiting for processing. Default is false.
speed
float
No
Speech Rate: Default is 1.0.
batch_size
string
No
Batch Parallelism: Default is 1. Increasing this parameter can effectively speed up processing (e.g., setting it to 10), but it will also occupy more GPU resources. Note: This may involve additional costs.
UID: User Identification
Session: Session ID obtained after login
Example
JSON Payload Format
(1) All Parameters
(2) Minimum Parameters
Example
Last updated
Was this helpful?