This weekend I attempted to set up a serveless environment to process files generated from the voice recorder. I got hands on experience on AWS IAM, Lambda, S3, bucket policies etc. I am still not sure about the best architecture. I think that the voice recordings can be uploaded to S3 and lambda function can process these files to report the voice pitch.
I prepared a python file that can detect the pitch, I couldnt figure out if this is possible on the website via pure js. Running python on the browser (brython and especially pyodide as I do need numpy and Pandas libraries among others) is a topic I explored the least .
Any ideas, feedback is welcome.