Skip to content

feat(vertexai): Add flutter_soloud for sound output in Live API audio streaming example. #17305

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 7 commits into
base: main
Choose a base branch
from

Conversation

khanhnwin
Copy link

Description

Replace this paragraph with a description of what this PR is doing. If you're modifying existing behavior, describe the existing behavior, how this PR is changing it, and what motivated the change.

Related Issues

Replace this paragraph with a list of issues related to this PR from the issue database. Indicate, which of these issues are resolved or fixed by this PR. Note that you'll have to prefix the issue numbers with flutter/flutter#.

Checklist

Before you create this PR confirm that it meets all requirements listed below by checking the relevant checkboxes ([x]).
This will ensure a smooth and quick review process. Updating the pubspec.yaml and changelogs is not required.

  • I read the Contributor Guide and followed the process outlined there for submitting PRs.
  • My PR includes unit or integration tests for all changed/updated/fixed behaviors (See Contributor Guide).
  • All existing and new tests are passing.
  • I updated/added relevant documentation (doc comments with ///).
  • The analyzer (melos run analyze) does not report any problems on my PR.
  • I read and followed the Flutter Style Guide.
  • I signed the CLA.
  • I am willing to follow-up on review comments in a timely manner.

Breaking Change

Does your PR require plugin users to manually update their apps to accommodate your change?

  • Yes, this is a breaking change.
  • No, this is not a breaking change.

@MichaelVerdon
Copy link
Contributor

MichaelVerdon commented Apr 24, 2025

Hi there, thanks for the PR! I understand that this is a draft but can you please give me some more detail about this? I recently added a sound example to vertexai https://github.com/firebase/flutterfire/blob/main/packages/firebase_vertexai/firebase_vertexai/example/lib/pages/audio_page.dart unless yours is different?

@khanhnwin
Copy link
Author

khanhnwin commented Apr 24, 2025

Hey @MichaelVerdon! From my understanding, the audio page that you linked to is for recording a it of audio and sending it to the Gemini API as opposed to streaming live audio, right?

I'm working with @cynthiajoan on refining the Live API example page, so in this case it's bidirectional audio streaming here => https://github.com/firebase/flutterfire/blob/main/packages/firebase_vertexai/firebase_vertexai/example/lib/pages/bidi_page.dart. Docs: https://firebase.google.com/docs/vertex-ai/live-api?platform=flutter

Edit: I realize my PR title was unclear, so I added more detail!

@khanhnwin khanhnwin changed the title Add flutter_soloud for sound output example. Add flutter_soloud for sound output in Live API audio streaming example. Apr 24, 2025
@cynthiajoan cynthiajoan changed the title Add flutter_soloud for sound output in Live API audio streaming example. feat(vertexai): Add flutter_soloud for sound output in Live API audio streaming example. Apr 24, 2025
@russellwheatley
Copy link
Member

@khanhnwin - is this still being worked on?

@khanhnwin
Copy link
Author

@russellwheatley Yes! @cynthiajoan is going to clean it up and integrate it into the example app is my understanding.

@cynthiajoan
Copy link
Collaborator

@russellwheatley Yes! @cynthiajoan is going to clean it up and integrate it into the example app is my understanding.

Yeah I'm working on this and not planning to have this PR included in the I/O release.

@kevsjh
Copy link

kevsjh commented May 24, 2025

The current implementation of Live AI API stream gets the AI to enter an endless loop of self replying, the audio of the speaker feeds back into the user audio stream due to the lack of echo cancelling.

Tested against both flutter_soloud and just_audio where both packages does not have AEC implementation so far.

What is the recommended flutter audio package for this use case as currently this approach is not usable for production

@cynthiajoan cynthiajoan marked this pull request as ready for review May 30, 2025 23:03
@cynthiajoan
Copy link
Collaborator

The current implementation of Live AI API stream gets the AI to enter an endless loop of self replying, the audio of the speaker feeds back into the user audio stream due to the lack of echo cancelling.

Tested against both flutter_soloud and just_audio where both packages does not have AEC implementation so far.

What is the recommended flutter audio package for this use case as currently this approach is not usable for production

In this PR we updated the audio_input.dart to add some attempts for AEC in the recorder config. Would you like to try to see if that improves the AI self hearing in your target platform?

Comment on lines +52 to +53
final AudioOutput audioOutput = AudioOutput();
final AudioInput audioInput = AudioInput();
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can these be kept private? I don't see any references outside this library.

Also consider omitting the repeated type name.

Suggested change
final AudioOutput audioOutput = AudioOutput();
final AudioInput audioInput = AudioInput();
final _audioOutput = AudioOutput();
final _audioInput = AudioInput();

initAudioOutput();
}

Future<void> initAudioOutput() async {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can this be private? Consider renaming _initAudio since it handles both input and output

});
await _session.sendMediaStream(mediaChunkStream);
if (inputStream != null) {
Stream<InlineDataPart> inlineDataStream = inputStream.map((data) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[nit] Consider omitting the variable type on the left.

Suggested change
Stream<InlineDataPart> inlineDataStream = inputStream.map((data) {
final inlineDataStream = inputStream.map((data) {

} catch (e) {
print(e);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this an intended change? Consider removing all print. Maybe use log instead?

throw UnimplementedError(
'Function not declared to the model: ${functionCall.name}',
);
for (final functionCall in functionCalls) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks like a fine change to make - how does it relate to the rest of the PR? Consider comiting separately.

super.dispose();
}

Future<void> checkPermission() async {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can this be private?

SoundHandle? handle;

Future<void> init() async {
/// Initialize the player.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[nit] use // for comments that are not member docs.

Suggested change
/// Initialize the player.
// Initialize the player.

import 'package:flutter_soloud/flutter_soloud.dart';

class AudioOutput {
AudioSource? stream; // Start playback
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't understand this comment. Should it be expanded?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants