Speech to text API not returning long running operation results - is service impaired?

Google Cloud Build with AI AI APIs
1

The Speech to text API stopped returning results today. It was working properly yesterday evening (16 hours ago.

I have tried it from python and node, both of which worked fine yesterday.

I am using the same code and short audio (2 minutes) that I have been using.

I am using the python speech_v1p1beta1 lib and have tried speech_v1 also.

config = {
    'encoding': 'FLAC',
    'language_code': 'en-US',
    'audio_channel_count': channel_count,
    'sample_rate_hertz': sample_rate,
    'enable_word_time_offsets': True,
    'enable_separate_recognition_per_channel': False,
    'enable_automatic_punctuation': True,
    'use_enhanced': True,
    'model': 'video'
}
audio = {
    'uri': audio_uri
}
request = {
    'config': config,
    'audio': audio
}
operation = cls.client.long_running_recognize(request)
response = operation.result(timeout=10000)
transcript = cls.build_transcript(response)

I get the operation object but the results just timeout no matter what timeout I set

Thanks

2

It is working again as of late yesterday but timestamps are off.

3

It looks like there’s no recent service impairment of Speech-to-Text. If there are any ongoing incidents, you should see them listed on the dashboard. As for the timestamp, you may want to check this documentation.