Creating a Voice Training Task (Advanced Edition)

Function

This API is used to create a voice training task (advanced edition). This API returns an OBS URL for uploading voice files.

Voice files can be uploaded in either of the following ways:

Compressing voice files and text files into a .zip file for upload: A voice file has been divided into 100 .wav files. Each voice file corresponds to a .txt file. All files are compressed into a .zip file. Voice files are named in the format of x.wav (x is an integer ranging from 0 to 99). Text files are named in the format of x.txt (x is an integer ranging from 0 to 99).
Uploading voice files and text files sentence by sentence: Upload the voice file and text file of a sentence each time, and then call the API for confirming the online recording result to check whether the voice is consistent with the text. After the confirmation, upload and confirm the next sentence.

After the files are uploaded, call the API for submitting a voice training task to start review and training.

Calling Method

For details, see Calling APIs.

URI

POST /v1/{project_id}/voice-training-manage/user/middle-jobs

**Table 1** Path Parameters
Parameter	Mandatory	Type	Description
project_id	Yes	String	Project ID. For details about how to obtain the project ID, see Obtaining a Project ID.

Request Parameters

**Table 2** Request header parameters
Parameter	Mandatory	Type	Description
X-Auth-Token	No	String	User token. This parameter is mandatory when token authentication is used. You can obtain the token by calling the IAM API used to obtain a user token. Value of X-Subject-Token in the response header.
Authorization	No	String	Authentication information. This parameter is mandatory for AK/SK authentication.
X-Sdk-Date	No	String	Time when the request is sent. This parameter is mandatory for AK/SK authentication. The format is YYYYMMDD'T'HHMMSS'Z'.
X-Project-Id	No	String	Project ID. This parameter is mandatory for AK/SK authentication.
X-App-UserId	No	String	Third-party user ID, which does not allow Chinese characters.

**Table 3** Request body parameters
Parameter	Mandatory	Type	Description
tag	No	String	Task tag. ECOMMERCE NEWS MARKETING
description	No	String	Description, which is displayed in the asset library.
sex	No	String	Voice gender, which can be male or female. FEMALE MALE
voice_name	Yes	String	Timbre name. The name is used as the name of the timbre model asset in the asset library.
language	No	String	Training language. Currently, only Chinese is supported. CN: Chinese EN: English
create_type	No	String	Task creation mode. PACKAGE: A .zip file contains all data. SEGMENT: Data is uploaded sentence by sentence.
phone	No	String	Mobile phone number

Response Parameters

Status code: 200

**Table 4** Response body parameters
Parameter	Type	Description
job_id	String	Task ID.
training_data_uploading_url	String	URL of the uploaded training data. You need to compress the training data into a .zip file and upload the file to this URL. This parameter is specified when create_type is set to package. NOTE: Set content-type to application/zip when uploading files to the OBS URL.
segment_uploading_url	segment_uploading_url object	URL of an upload task sentence by sentence. This parameter is specified when create_type is set to segment.
authorization_letter_uploading_url	String	Upload URL of the authorization letter.

**Table 5** segment_uploading_url
Parameter	Type	Description
audio_uploading_url	Array of strings	URL for uploading audio files. NOTE: Set content-type to audio/wav when uploading files to the OBS URL.
txt_uploading_url	Array of strings	URL for uploading text files. NOTE: Set content-type to text/plain when uploading files to the OBS URL.

Status code: 400

**Table 6** Response body parameters
Parameter	Type	Description
error_code	String	Error code.
error_msg	String	Error description.

Example Requests

POST https://{endpoint}/v1/3f0924078d1b471c884a5383d4dec9fa/voice-training-manage/user/middle-jobs

{
  "tag" : "ECOMMERCE",
  "description" : "It is a female voice.",
  "sex" : "FEMALE",
  "voice_name" : "Gentle female voice",
  "language" : "CN",
  "create_type" : "PACKAGE"
}

Example Responses

Status code: 200

The information is returned when the request succeeded.

{
  "job_id" : "abcdefg",
  "training_data_uploading_url" : "https://my-bucket/data.zip",
  "segment_uploading_url" : {
    "audio_uploading_url" : [ "https://my-bucket/data0.wav" ],
    "txt_uploading_url" : [ "https://my-bucket/data0.txt" ]
  },
  "authorization_letter_uploading_url" : "https://my-bucket/data"
}