I am planning to use Google Speech-To-Text to transcribe around 14,000 radio shows (.flac files with one channel), each around 15 Min long. All data was uploaded to a Google Storage Bucket. I planned to transcribe the shows via the API, but given the around 145 days of content, obtaining the information would take more than a month.
I was wondering: are there any ways to speed up the process, i.e. to send parallel requests or similar?
I'm a newbie to Google Cloud services, so any advice is highly appreciated.
As mentioned by @Jonskeet, you can create multiple long running requests using client library and store the results in GCS bucket following the documentation.
With regards to quota and limits, since your input audio is in GCS there is no size limit on requests. However, you will have to look into Google Speech-to-Text API's request limits mentioned in this documentation for quota limits.