Greasy Fork

LingQ Azure TTS Generator

Add the ability to generate Azure Text-to-Speech (TTS) to LingQ.

当前为 2023-04-20 提交的版本，查看 最新版本。

作者
Kevin (GkevinOD)
评分
0 0 0
版本
1.1
创建于
2023-04-18
更新于
2023-04-20
大小
18.1 KB
许可证
GPL-3.0-or-later
适用于

This program adds a "Generate Audio" button to the lesson editor tab, enabling users to generate audio using Azure Neural Text-to-Speech (TTS) and create corresponding timestamps for each sentence. To use this feature, you will need a key and service region from the Microsoft Azure Speech Service.

How to obtain the key and service region

  1. If you don't have a Microsoft Azure account, sign up for one here.
  2. Once logged in, navigate to the Azure portal.
  3. In the left-hand menu, click on Create a resource.
  4. Search for "Speech" in the search box and select Speech from the results.
  5. Click the Create button to create a new Speech Service resource.
  6. Fill in the required details, such as subscription, resource group, name, region, and pricing tier.
  7. Click Review + create and then Create to finalize the creation process.
  8. After the resource deployment is complete, go to the resource's Overview page.
  9. Note down the Key and Location (region), which you will need for the program.

Speech voices list

For the list of available speech voices, please refer to the Microsoft Azure Cognitive Services Speech Service Language Support documentation.

Using the program

  1. Open the lesson editor tab and click on the Generate Audio button.
    • If the "Generate Audio" button doesn't show up, refresh the page and try again.
  2. A prompt box will appear, requesting input for the speech voice, key, and region.
  3. Enter the speech voice (e.g., en-US-NancyNeural), key, and region obtained from your Azure Speech Service resource.
  4. Choose whether to Cancel the operation or Generate with the audio generation.
  5. If you proceed, a progress bar will display the current progress of audio generation, keeping you informed of the status as the audio is being created.