Skip to content

语音识别本地文件的时候每次都要加载模型 #3867

Open
@Alex86706

Description

@Alex86706

识别一个简短的音频也需要4s多,原因是加载模型需要这么长时间,为什么这个不能像其他的算法框架一样,可以预加载模型,然后直接去识别文件呢?

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions