Skip to content

Accessing ASR module to stream transcript #167

Description

@KarthiU

I’m currently working on functionality that would allow the robot to trigger a specific chain of commands through voice input. I’m running into some limitations with the ASR module and wanted to ask whether there’s a way to access the live audio transcript directly, without going through the full wake-up/conversation mode flow.

Ideally, I’d like the system to simply stream the recognized text locally, without prompting me to connect to the internet or requiring wake-up conversation mode to be enabled. Right now, although streaming is place, it keeps triggering the cloud access loop (asks user to connect to the internet, and changes LED colors), which inhibits my use case

Is this flow customizable in general? Specifically, is there a supported way to communicate with the robot by voice without enabling the full wake-up conversation mode?

For context, I’m using a Unitree G1 EDU+.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions