Skip to main content
POST
For the US West endpoint, see Synchronous Text to Speech · Endpoints.

Authorizations

​
Authorization
string
header
required

HTTP: Bearer Auth

Headers

​
Content-Type
enum<string>
default:application/json
required

The media type of the request body. Must be set to application/json to ensure the data is sent in JSON format.

Available options:
application/json

Body

application/json
​
model
enum<string>
required

The speech synthesis model version to use.

Available options:
speech-2.8-hd,
speech-2.8-turbo,
speech-2.6-hd,
speech-2.6-turbo,
speech-02-hd,
speech-02-turbo,
speech-01-hd,
speech-01-turbo
​
text
string
required

The text to be converted into speech. Must be less than 10,000 characters.

  • For texts over 3,000 characters, streaming output is recommended.

  • Paragraph breaks should be marked with newline characters.

  • Pause control: You can customize speech pauses by adding markers in the form <#x#>, where x is the pause duration in seconds. Valid range: [0.01, 99.99], up to two decimal places. Pause markers must be placed between speakable text segments and cannot be used consecutively.

  • Inline pronunciation: Wrap Mandarin Pinyin (with tone number 1–5) or IPA symbols or Cantonese Jyutping (with tone number 1–6) in half-width parentheses to override pronunciation of the target word or polyphonic character.

    • "The word live is pronounced (lɪv) as a verb and (laɪv) as an adjective."
    • "This is (he2)平, not (huo4)面."
    • "去街市買啲(sung3)。"

  • Interjection tags: Only supported when using speech-2.8-hd or speech-2.8-turbo models. Supported interjections: (laughs), (chuckle), (coughs), (clear-throat), (groans), (breath), (pant), (inhale), (exhale), (gasps), (sniffs), (sighs), (snorts), (burps), (lip-smacking), (humming), (hissing), (emm), (sneezes).

​
stream
boolean

Whether to enable streaming output. Defaults to false.

​
stream_options
object
​
voice_setting
object
​
audio_setting
object
​
pronunciation_dict
object
​
timbre_weights
object[]

Timbre weights (legacy field)

​
language_boost
enum<string>

Controls whether recognition for specific minority languages and dialects is enhanced. Default is null. If the language type is unknown, set to "auto" and the model will automatically detect it.

Note: The speech-01 and speech-02 series models do not currently support Persian, Filipino, or Tamil.

Available options:
Chinese,
Chinese,Yue,
English,
Arabic,
Russian,
Spanish,
French,
Portuguese,
German,
Turkish,
Dutch,
Ukrainian,
Vietnamese,
Indonesian,
Japanese,
Italian,
Korean,
Thai,
Polish,
Romanian,
Greek,
Czech,
Finnish,
Hindi,
Bulgarian,
Danish,
Hebrew,
Malay,
Persian,
Slovak,
Swedish,
Croatian,
Filipino,
Hungarian,
Norwegian,
Slovenian,
Catalan,
Nynorsk,
Tamil,
Afrikaans,
auto
​
text_normalization_mode
enum<string>
updated

Text normalization. basic uses fast, rule-based normalization. quality uses a hybrid LLM + rule-based approach to improve normalization quality, with additional latency. See Text Normalization.

Note:

  • quality currently takes effect only for non-streaming requests (stream=false) with speech-2.8-hd, speech-2.8-turbo, speech-2.6-hd, and speech-2.6-turbo. For streaming requests or other models, quality is processed as basic and no error is returned.
  • When this parameter is set, voice_setting.text_normalization is ignored. If it is omitted, the voice_setting.text_normalization setting applies (disabled by default).
Available options:
basic,
quality
​
voice_modify
object

Voice effects configuration.

Supported audio formats:

  • Non-streaming: mp3, wav, flac
  • Streaming: mp3
​
subtitle_enable
boolean
default:false

Controls whether subtitles are enabled. Default is false. Available for models: speech-2.8-hd, speech-2.8-turbo, speech-2.6-hd, speech-2.6-turbo, speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo.

​
subtitle_type
enum<string>
default:sentence

Subtitle granularity. Default is sentence. Options:

  • sentence: sentence-level timestamps
  • word: word-level timestamps
  • word_streaming: word-level timestamps optimized for streaming, only valid when stream=true

In streaming mode (stream=true), subtitles are delivered in data.subtitle along with the audio, and subtitle_file is not returned:

  • sentence / word: returned once per segment, on the last audio chunk of that segment. sentence does not include timestamped_words
  • word_streaming: returned on every audio chunk, carrying the cumulative subtitle of the current segment so far (timestamped_words and time_end keep growing). Replace the previous result of the same segment with the latest one; a change of text_begin means a new segment has started

In non-streaming mode, subtitles are returned as a file via subtitle_file.

Available options:
sentence,
word,
word_streaming
​
output_format
enum<string>
default:hex

Controls the output format. Options: [url, hex]. Default is hex. Only effective in non-streaming scenarios. In streaming, only hex is supported. Returned url is valid for 24 hours.

Available options:
url,
hex

Response

​
data
object

The synthesized audio data object. The returned data object may be null, so a null check is required.

​
trace_id
string

The session ID, used for troubleshooting and support.

​
extra_info
object

Additional audio information.

​
base_resp
object

Status code and details of this request.