DocsICP FilingConsole
官方文档
首页

Create a project

更新时间：
复制 MD 格式
产品详情
我的收藏

In the Intelligent Speech Interaction console, you can upload corpora, train models, and publish them to improve recognition accuracy. This topic describes how to create a project.

Prerequisites

The Intelligent Speech Interaction service has been activated. For more information, see Activate Service.

Procedure

  1. Log on to the Intelligent Speech Interaction console.

  2. By default, the overview page appears. In the left-side navigation pane, click All Projects.

  3. On the My All Projects page, click Create Project.

  4. In the Create Project dialog box, enter a project name, select a project type, and enter a project scenario description.

    The available project types are Speech-to-Text + Text-to-Speech + Speech Analytics, Speech-to-Text Only, Text-to-Speech Only, and On-device Solution.

    Type

    Description

    Speech-to-Text Only

    In the Speech-to-Text section, click Modify Configurations.

    For your use case, run an online test and click Confirm. In the Speech-to-Text model selection page that appears, filter by scenario and select your target model, such as Chinese Mandarin (Shiyinshi V1 - End-to-End Model), and click Confirm.

    Text-to-Speech Only

    To configure the Text-to-Speech model, first select a voice in the Select a Text-to-Speech model section by using the category tabs on the left (General, Customer Service, Livestreaming, Children's Voice, English, Dialect, and Multilingual) to filter and select a target voice, for example, Wencheng/Standard Male Voice. Then, in the Basic Parameters section, adjust the speech rate (from –500 to 500), pitch (from –500 to 500), and volume (from 0 to 100). Select a sample rate (48000, 24000, 16000, or 8000) and a format (mp3 or wav). After finishing the configuration, test the service and click Confirm.

    Speech-to-Text + Text-to-Speech + Speech Analytics

    This type provides all features of Intelligent Speech Interaction, which you can configure as needed.

    On-device Solution

    You must select an SDK type. The supported types are On-device Solution SDK, Standard Offline Text-to-Speech SDK, and Premium Offline Text-to-Speech SDK. In the Create Project dialog box, enter the project name, set the project type to On-device Solution, select the required SDK from the SDK type drop-down list, enter the project scenario description, and click OK.

  5. After the project is created, it appears on the My All Projects page.

Next steps

  • For projects of the Speech-to-Text + Text-to-Speech + Speech Analytics, Speech-to-Text Only, or Text-to-Speech Only types: Configure a project.

  • For On-device Solution projects: Run the SDK.