Overview
The voice/sound separation feature separates the voice from the background sound in a video material to generate a new independent audio file. Then, you can apply artistic processing of other styles to the material without accompaniment and noise.
Use Cases
Post audio mixing
Diverse stylistic and artistic processing operations can be performed on voices, background sounds, and accompaniments after separation, such as voice changing and audio mixing.
Voices can be quickly separated, and then different background sounds can be added to mix materials for targeted marketing to user groups with different interests.
Directions
You can use the voice/sound separation feature through jobs or workflows. In order to improve the operational efficiency and reduce repetitive operations, CI offers the template feature, which is a configuration item in jobs and workflows. You can save common parameter combinations as templates and reuse them directly in subsequent operations, with no need to set the parameters every time you start a job. You can customize voice/sound separation templates as follows:
Through job
You can create a voice/sound separation job through the console or API for existing data stored in COS.
Console: You can create a voice/sound separation job visually in the CI console as instructed in Job. Through workflow
CI provides the workflow service. You can enable a workflow for a bucket or a specific path. Then, voice/sound separation will be automatically performed on files uploaded to the bucket or path, and the output audio files will be saved in the specified location.
Creating workflow
You can create a workflow in the CI console as instructed in Workflow. Deleting, querying, and searching for workflow