Trying to do multiple voice files with speech-to-text

[CloudyGoogle]

Hello.

I'm someone who's trying to make speech-to-text work without being a coder in any way whatsoever. I have let's say hundreds of individual audio files and they go from 30 seconds to a minute and a half. The problem is that uploading them to the bucket makes it so there's hundreds of individual ones. And I need to create a transcriptions individually. what do I do? can I not just transcribe everything in one folder?

[anil789]

It requires coding - you load files in a bucket (say in 'input' folder) and run a background job to produce a "txt" file for each using speech-to-text API (say in 'output' folder).  If you have files formats such as mp4 then use transcoding. This is the step roughly.

[CloudyGoogle]

Understood. But do you have any idea how I'd code that? Of course I am  unfamiliar with how to code... but anyway no problem at all if you cannot help there.

[anil789]

Coding it will not be easy if you are not a programmer. It requires a bunch of technology and tools. Roughly steps would be:

1. Using gsutil tool of the GCP, upload files to a bucket.

2. Write a program to read file from the bucket and invoke  Speech-to-Text API. It will require you to acquire an access-token (OAuth2). 

3. If files are small then you could do 2 without uploading files to the bucket. 

You can find the example programs here: https://cloud.google.com/speech-to-text/docs/samples