fix(speechmatics): resample audio to the 16 kHz Agent STT accepts

[omChauhanDev]

Fixes #5717

Problem

Since 1.10.0 SpeechmaticsSTTService targets Agent STT (/v2/agent), which accepts 16 kHz only. The service still declared the pipeline's input rate in audio_format, so any pipeline not running at 16 kHz opens a session the server rejects with audio_format.sample_rate does not match: 16000. That includes the 8 kHz telephony setup pipecat init generates. The rejection is permanent, so the service is left unusable, and a Speechmatics instance sitting behind a ServiceSwitcher as a backup dies at pipeline start before it is ever selected.

Passing sample_rate=16000 to the service did not help: it changed the declared rate without converting the audio, so 8 kHz PCM went out labelled as 16 kHz.

Changes

run_stt now resamples audio to 16 kHz before sending it, and audio_format declares that fixed rate instead of the pipeline's, so the declared rate and the bytes cannot disagree. The conversion reads the rate recorded in setup(), which means a sample_rate override cannot declare a rate the audio does not have. Pipelines already running at 16 kHz are unaffected.

[codecov]

Codecov Report

❌ Patch coverage is 85.71429% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/pipecat/services/speechmatics/stt.py	85.71%	2 Missing ⚠️

Files with missing lines	Coverage Δ
src/pipecat/services/speechmatics/stt.py	73.39% <85.71%> (+2.16%)	⬆️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.