Repository navigation
Speech API faster with higher sample_rate #3161
Description
The function sync_recognize() of the Speech API has consistently a shorter runtime, if I use a flac audio file with a sampling rate of 48 kHz in comparison to a sampling rate of 16 kHz.
How does it come that the larger file with more information to process, is analyzed faster?
And as a follow-up question: The docs say that a sampling rate of 16 kHz yields the best results. What is the reason for that?
(I already asked the question on Stack Overflow with some code included, but got no answer so far.)
Activity
daspecster commentedon Mar 17, 2017
Thanks for the question @elcombato! To be totally honest, I'm not sure. I'm guessing it's just some optimization upstream.
@gguuss do you have any insight into this?
lukesneeringer commentedon Mar 17, 2017
Hi @elcombato,
This repository is for the client library, not the API itself. To be honest, I am not sure why you are getting distinct behavior there. You can feel free to follow-up with the API's support channels (as it seems you may have already done).
chore(deps): update dependency google-cloud-logging to v1.15.0 [(#3161)…