Help Center/ MetaStudio/ Getting Started/ Voice Modeling
Updated on 2025-08-25 GMT+08:00
View PDF
Share

Voice Modeling

MetaStudio allows customizing virtual avatar voices.

Prerequisites

Before creating a voice modeling task, you need to prepare the following items by referring to Procedure:

Video

Watch this video to learn how to train your voice model and create a lifelike voice for your virtual avatar.

Procedure

  1. Log in to the MetaStudio console and go to the Overview page.
  2. Click Go to MetaStudio Console to go to the MetaStudio console.
  1. Click the Voice Modeling card to go to the voice modeling page.

    Figure 1 Customizing a voice

  2. Under the Huawei Models tab, configure voice modeling parameters.

    For details, see Table 1.
    Table 1 GUI operations

    Parameter

    Description

    Voice modeling

    Select Voice modeling and record a WAV or MP3 audio, with a pause of 2–3 seconds between phrases. The audio duration ranges from 10–30 minutes (recommended: 15 minutes).

    Voice Settings

    Enter a voice name. Example: joyful female

    Produce Voice

    If you select Script Upload, upload a recorded WAV or MP3 audio. You do not need to compress the audio or upload TXT files.

    Voice Gender

    Gender of the voice. Example: Female

    Input Language

    Select the language of the uploaded script, for example, Chinese.

    Voice Field

    Field to which a voice applies. Select a field based on the selected script example.

  3. Check the box for authorizing the voice use and click Submit.

    The Information dialog box is displayed, notifying you of the remaining voice modeling quota and indicating that one resource will be consumed this time.

  4. After confirming the information, click Submit. After the voice modeling task is submitted, the message Production task submitted is displayed.

    Model review and modeling take about seven working days.

  5. You can click View Production Tasks to view the review progress of the voice modeling task.

    When the status changes to Reviewed, algorithm training is automatically started. If there are multiple algorithm training tasks, queuing and delay may occur.

  6. After the training is complete, choose Assets > My Models in the navigation pane.
  7. Select the Voices > Activated tab, find the generated voice, and click the avatar in the voice card to preview the voice.

    Figure 2 Voice