commit ca986b96d65f3adcb4a5a8a5934b10bc0983ac51
author avynn <avynn@google.com> Mon Aug 26 16:20:11 2024
committer Chromium LUCI CQ <chromium-scoped@luci-project-accounts.iam.gserviceaccount.com> Mon Aug 26 16:20:11 2024
tree 0fdf10a7655ac8e99deb9699e372252c5aa97dd7
parent 0eb200c0678d7622a855d50bfa379a1be126cd75

[BabelOrca] FakeSpeechRecognitionService supports multiple recognizers

Re-factors the Fake SpeechRecognitionService to support multiple
recognizers or AudioSourceFetchers.  This is done by re-factoring the
Service such that tests will assert against specific recognizers
instead of the service as a whole.

Justification for this re-factor comes from three sources:

1.) Since this test dependency is used in browser tests it should
behave as much like the real speech recognition service as possible.
This includes the fact that the real service can handle multiple
recognizers concurrently.

2.) In the next CL we will implement the majority of
go/babel-orca-mic-integration which modifies the
SystemLiveCaptionService.  In the new paradigm we want to be able
to test in an environment in which both versions of the
SystemLiveCaptionService are instantiated at the same time. see
the design for more information.

3.) In the future we plan on explicitly supporting mutliple
concurrent recognizers for LiveCaption.  In order to test behavior
there we have to ensure that tests can also support mutliple
concurrent recognizers.

DanglingUntriaged-notes: Adding new dangling pointers in tests only
AX-Relnotes: N/A
Design: go/babel-orca-mic-integration
Bug: N/A
Change-Id: I33353ead7a1dca48b630d9fd0e4386035ca63c2b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5664296
Reviewed-by: Devlin Cronin <rdevlin.cronin@chromium.org>
Reviewed-by: Evan Liu <evliu@google.com>
Reviewed-by: Ahmed Nasr <anasr@google.com>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Katie Dektar <katie@chromium.org>
Commit-Queue: Avynn Donaghe <avynn@google.com>
Cr-Commit-Position: refs/heads/main@{#1346771}