Need support inputAudioTranscription in core/src/main/java/com/google/adk/agents/RunConfig.java

[shenshaoyong]

** Please make sure you read the contribution guide and file the issues in the right place. **
Contribution guide.

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
Need to transcribe input audio
Just like in https://github.com/googleapis
java-genai

https://github.com/googleapis/java-genai/blob/main/src/main/java/com/google/genai/types/LiveConnectConfig.java

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.
waiting

Additional context
Add any other context or screenshots about the feature request here.

[MinkeCheng]

Title: Feature Request: Add inputAudioTranscription Support in Java ADK

Description:
I have also encountered this requirement and noticed that the inputAudioTranscription parameter, which converts input audio to text transcription, is available in the Python ADK. This feature would be very useful for processing and analyzing audio input in real-time conversations.

Currently, while the Java ADK supports audio input/output streaming (as seen in LiveAudioRun implementation), it lacks the ability to automatically transcribe the input audio to text.

Request:

• Please consider adding inputAudioTranscription parameter support to the Java ADK
• This would bring feature parity with the Python ADK implementation
• It would greatly enhance the capabilities of audio processing in Java applications

Use Case:

• Real-time audio transcription during conversations
• Text analysis of audio input
• Logging and monitoring of audio conversations

Would appreciate if this feature could be prioritized for an upcoming release.

[huangduxin]

+1 for this feature!

I'm also working with the Java ADK for audio streaming and noticed this limitation. The Python version has inputAudioTranscription which is super helpful for getting text from audio input, but can't find similar functionality in Java.

Looking at my LiveAudioRun implementation, we can capture and stream audio but there's no built-in way to get the text transcription. Would be really useful to have this in Java ADK too.

Any timeline for adding this feature? It would save us from having to implement workarounds.

[zhuqianyun-Rhine]

Bumping this thread!

Running into the same limitation in Java ADK. We really need the inputAudioTranscription feature that Python ADK has.

Quick context:
Running voice chat app using LiveAudioRun, everything works smooth except can't get text from audio input. In Python it's just one param but Java doesn't have it yet.

Example config we're using:

RunConfig.builder()
    .setStreamingMode(RunConfig.StreamingMode.BIDI)
    .setResponseModalities(ImmutableList.of(new Modality("AUDIO")))
    // <- would be great to have inputAudioTranscription here

Having this would help with:

• Debug logs
• User analytics
• Content moderation

Anyone else need this? Or any workarounds for now?

[vorburger]

@shenshaoyong @MinkeCheng @huangduxin @zhuqianyun-Rhine if you would like to raise any PRs to make contributions for this, even (and better) as small incremental steps towards this, we would certainly very much welcome engagement with pull requests suggesting improvements from engaged community members!

[jinnigu]

@vorburger @shenshaoyong I would like to implement this feature. Please take a look at this PR: #463

[Poggecci]

PR #463 was merged