Skip to content

Fun-ASR-MLT-Nano模型效果很差 #32

@zyjcsf

Description

@zyjcsf

我想请问下,我用这个模型解码librispeech的clean和other测试集wer分别是25.43%,24.84%是否合理,我看存在很多后半段漏识别的问题;解码参数均是默认参数

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