Skip to content

Can be used for speech clustering #65

Description

@MirMagchw

Hello, I would like to ask a question. I have the recording data of 5000 speakers, and each person has multiple recording segments. I use a neural network to extract features from the speech, the speaker identity is used as the label, and a clustering network is trained using the method provided by you. Then use this clustering network to extract features from the audio clips of other people and cluster them. Is it feasible in theory? Thank you very much for your answer

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions