How to Use the ByteDance Speech Synthesis Large Model 2.0 Channel

Please note that ByteDance has many speech synthesis products with similar names. This channel corresponds specifically to Speech Synthesis Large Model --> Large Model Speech Synthesis API within ByteDance's offerings. Please ensure accuracy, otherwise it will not work.

How to Activate

First, register, log in, and complete real-name verification.

https://console.volcengine.com/

Open this link to register, log in, and complete real-name verification.

Enter the backend to create an application: The ByteDance backend is cluttered with many similarly named menu items. Please make sure to navigate to -- Doubao Speech Synthesis Model 2.0. Follow the image below to enter.

Or directly click this link: https://console.volcengine.com/speech/app

Note: Please switch to Legacy Version in the top-left corner. The new version is too messy and difficult to navigate.

As shown in the image below -- Legacy Version -- Application Center -- Application Management -- Create Application.

You can fill in the name and description arbitrarily, but crucially, you must check the box for Doubao Speech Synthesis Model 2.0--Doubao Speech Synthesis Model 2.0 Character Version. There are many similarly named Speech Synthesis options here, so be careful not to select the wrong one.

Activate a trial or formal service: Next, go to the Speech Synthesis page to activate the free trial.

Navigate to: https://console.volcengine.com/speech/service/10035

At the top, select the application you just created and click "Trial" to activate it.

Copy information for use: Copy 3 parameters, then you can fill them into the video translation software for use.

The 1st is App id. Scroll down on the same page to find it.

The 2nd is Access Token, located to the right of App id. Copy it.

Fill them into the video translation software. Open Menu - TTS Settings -- ByteDance Speech Synthesis Model 2.0 window, fill in the details, test to ensure no issues, then save.

Using it in the Video Translation Software

After filling in and testing successfully, first select the target language in the software, then choose ByteDance Speech Synthesis Model 2.0 in the dubbing channel. You can click to preview each voice.

Select a satisfactory voice to start the dubbing operation.

Important Note

Many voices require you to activate and purchase them separately in the ByteDance backend. If you encounter failures or "unavailable" messages when using a specific voice, it most likely needs to be purchased separately.