Create Speech Generation Task

Returned File Information

The return result for a single file input is shown below.
If the input is a compressed package containing multiple files, a corresponding folder will be generated for each file. The contents inside each folder are the same as those for a single file input.

Input File Type: txt File

Output Files:
- Audio File: Format follows the request body settings.
- Subtitle File: Sentence-level subtitle information.
- Extra JSON File: Additional information related to the audio file.

Input File Type: json File

title Field Output Files (if this field is empty, no files will be generated)
- Audio File: Format follows the request body settings
- Subtitle File: Sentence-level subtitle information
- Extra JSON File: Additional information related to the audio file
content Field Output Files (if this field is empty, no files will be generated)
- Audio File: Format follows the request body settings
- Subtitle File: Sentence-level subtitle information
- Extra JSON File: Additional information related to the audio file
extra Field Output Files (if this field is empty, no files will be generated)
- Audio File: Format follows the request body settings
- Subtitle File: Sentence-level subtitle information
- Extra JSON File: Additional information related to the audio file

Authorizations

Authorization

string

header

required

HTTP: Bearer Auth

Security Scheme Type: http
HTTP Authorization Scheme: Bearer API_key, can be found in Account Management>API Keys.

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

Model version to call. Supported

Available options:

speech-2.6-hd,

speech-2.6-turbo,

speech-02-hd,

speech-02-turbo,

speech-01-hd,

speech-01-turbo

text

string

required

Text content to convert to audio, max length 50,000 characters. Mutually exclusive with text_file_id (one is required).

text_file_id

integer<int64>

required

ID of the text file to synthesize. Max 100,000 characters. Supported formats: txt, zip. Mutually exclusive with text (one is required).

txt file: Supports customizing 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.
zip file: Must contain files of the same type (txt or json).
- json format supports ["title", "content", "extra"] fields. Each non-empty field generates an audio file, subtitles, and metadata and would be stored in a folder.

voice_setting

object

required

Show child attributes

voice_setting.voice_id

string

required

The ID of the target voice.

To apply mixed voices, configure the timber_weights parameter and leave this value empty.
Supports system voices, cloned voices, and AI-generated voices. Below is a selection of the latest system voices (IDs). The full list of available voices can be viewed on the System Voice ID List or retrieved programmatically using the Get Voice API.
- Chinese:
  - moss_audio_ce44fc67-7ce3-11f0-8de5-96e35d26fb85
  - moss_audio_aaa1346a-7ce7-11f0-8e61-2e6e3c7ee85d
  - Chinese (Mandarin)_Lyrical_Voice
  - Chinese (Mandarin)_HK_Flight_Attendant
- English:
  - English_Graceful_Lady
  - English_Insightful_Speaker
  - English_radiant_girl
  - English_Persuasive_Man
  - moss_audio_6dc281eb-713c-11f0-a447-9613c873494c
  - moss_audio_570551b1-735c-11f0-b236-0adeeecad052
  - moss_audio_ad5baf92-735f-11f0-8263-fe5a2fe98ec8
  - English_Lucky_Robot
- Japanese:
  - Japanese_Whisper_Belle
  - moss_audio_24875c4a-7be4-11f0-9359-4e72c55db738
  - moss_audio_7f4ee608-78ea-11f0-bb73-1e2a4cfcd245
  - moss_audio_c1a6a3ac-7be6-11f0-8e8e-36b92fbb4f95

voice_setting.speed

number<float>

default:1

Speech speed. Higher values result in faster speech. Range: [0.5, 2] (default: 1.0).

Required range: 0.5 <= x <= 2

voice_setting.vol

number<float>

default:1

Speech volume. Higher values increase loudness. Range: (0, 10] (default: 1.0).

Required range: x <= 10

voice_setting.pitch

integer

default:0

Speech pitch adjustment. Range: [-12, 12] (default: 0, original pitch).

Required range: -12 <= x <= 12

voice_setting.emotion

enum<string>

Emotion control for synthesized speech. Supported values: ["happy", "sad", "angry", "fearful", "disgusted", "surprised", "calm", "fluent", "whisper"].

By default, the model automatically selects the most natural emotion based on text. Manual specification is only recommended when explicitly needed.
Available for models: speech-2.6-hd, speech-2.6-turbo, speech-02-hd, speech-02-turbo, speech-01-hd, speech-01-turbo.
Option fluent, whisper is only available for models: speech-2.6-turbo, speech-2.6-hd.

Available options:

happy,

sad,

angry,

fearful,

disgusted,

surprised,

calm,

fluent,

whisper

voice_setting.english_normalization

boolean

default:false

Enable text normalization for English. Improves performance in digit-reading scenarios at the cost of slightly higher latency. Default: false.

voice_setting.latex_read

boolean

default:false

Enable LaTeX formula reading.

Formulas must be wrapped with $$.
If the request contains a formula with "\", it must be escaped as "\\".

Example: The quadratic formula

should be written as $$x = \\frac{-b \\pm \\sqrt{b^2 - 4ac}}{2a}$$

audio_setting

object

Show child attributes

audio_setting.audio_sample_rate

integer<int64>

default:32000

Specifies the sampling rate of the generated audio. Supported values: [8000, 16000, 22050, 24000, 32000, 44100].

audio_setting.bitrate

integer<int64>

default:128000

Specifies the bitrate of the generated audio. Supported values: [32000, 64000, 128000, 256000]. Note: This parameter only applies to audio in mp3 format.

audio_setting.format

enum<string>

default:mp3

Specifies the format of the generated audio. Supported values: [mp3, pcm, flac] . Note: wav is only supported in non-streaming mode.

Available options:

mp3,

pcm,

flac

audio_setting.channel

integer<int64>

default:2

Specifies the number of audio channels. Supported values: [1, 2]. 1 = mono 2 = stereo Default is 1.

pronunciation_dict

object

Show child attributes

pronunciation_dict.tone

string[]

Defines pronunciation rules for specific characters or symbols.

For Chinese text, tones are represented numerically: 1 = first tone, 2 = second tone, 3 = third tone, 4 = fourth tone, 5 = neutral tone.

Example: ["omg/oh my god"]

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.

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

voice_modify

object

Voice effect settings. Supported formats: mp3, flac.

Show child attributes

voice_modify.pitch

integer

Corresponds to the "Deepen/Brighten" slider on the official page. Range: [-100, 100]. Values closer to -100 produce a deeper voice, while values closer to 100 result in a brighter tone.

Required range: -100 <= x <= 100

voice_modify.intensity

integer

Corresponds to the "Stronger/Softer" slider on the official page. Range: [-100, 100]. Values closer to -100 create a stronger, more forceful sound, while values closer to 100 yield a softer tone.

Required range: -100 <= x <= 100

voice_modify.timbre

integer

Corresponds to the "Nasal/Crisp" slider on the official page. Range: [-100, 100]. Values closer to -100 produce a fuller, richer sound, while values closer to 100 generate a crisper tone.

Required range: -100 <= x <= 100

voice_modify.sound_effects

enum<string>

Sound effects. Only one can be applied at a time. Options: spacious_echo, auditorium_echo, lofi_telephone, robotic

Available options:

spacious_echo,

auditorium_echo,

lofi_telephone,

robotic

Response

200 - application/json

task_id

string

Task ID

file_id

integer<int64>

The corresponding audio file ID is returned once the task is successfully created.

When the task is complete, you can use the file_id to call the File (Retrieve) API to download the file.

If the request fails, this field will not be returned.

Note: The download URL is valid for 9 hours (32,400 seconds) from the time it is generated. After expiration, the file will no longer be available and the generated data will be lost, so please ensure you download it within the validity period.

task_token

string

Token for completing the task

usage_characters

integer

Number of billed characters

base_resp

object

Status code and details.

Show child attributes

base_resp.status_code

integer<int64>

required

The status code.

0: Request succeeded
1000: Unknown error
1001: Timeout
1002: Rate limit exceeded
1004: Authentication failed
1042: More than 10% invalid characters
2013: Invalid input format

For more details, see Error Code Reference.

base_resp.status_msg

string

required

Detailed status message.

Text

Speech

Video

Image

Music

File

Error Code

Create Speech Generation Task

Returned File Information

Input File Type: txt File

Input File Type: json File

Authorizations

Headers

Body

Response

Text

Speech

Video

Image

Music

File

Error Code

​Returned File Information

​Input File Type: txt File

​Input File Type: json File

Authorizations

Headers

Body

Response

Returned File Information

Input File Type: txt File

Input File Type: json File