This topic describes how to configure projects for ASR and TTS.
Prerequisites
You have created a project. See Create a project.
Configure a project
On the All Projects page of the Intelligent Speech Interaction console, find your project and click Configure in the Actions column.
ASR
-
In the ASR module, click Modify Configurations.
-
In the scenario list, select a base model or add a custom model. For more information about how to use a custom model, see Manage custom models.
-
In the ASR module, you can configure hotwords and language models. Click Apply. For more information, see Model Customization.
TTS
In the TTS module, select a TTS model and configure the basic parameters (speech rate, pitch, and volume). After you publish the configuration, it will be bound to the project's Appkey. If you do not set these parameters in your application, it will use the default values from the console.
-
Select a voice. For more information about the use cases for different voices, see Speech Synthesis API reference.
-
In the Basic Parameters module, set the speech rate, pitch, and volume.
-
In the Test module on the right, preview the audio output.
-
Click Confirm. The configuration takes effect immediately.
On the TTS configuration page, first select a voice category from the left pane, such as General, Customer Service, Live Streaming, Child's Voice, English, Dialect, or Multilingual. Then, from the cards on the right, select a voice model, such as Siqi for a gentle female voice. In the basic parameters section, adjust the speech rate, pitch, and volume, and select a sample rate (48000/24000/16000/8000) and format (mp3/wav). In the Test area, you can enter text, click the speaker icon to preview the audio, or click the download icon to save the audio file.