I am trying to generate a transcription from Google's speech-to-text service from sample audio files. I tried some 1 min audio files, and it generated good transcription. But when I tied 30 mins or 1-hour files from some random podcast for testing, it failed 100% of the time. These are my configuration -
Language code: en-US
Transcription model: Latest Long
Word confidence: Enabled
Automatic punctuation: Enabled
Speaker diarization: Enabled
Min speaker count: 2
Max speaker count: 4
I will also attach some screenshots. Any solution, suggestion or comment will help. Thanks
