How to Use Nova Sonic Speech-to-Speech with Python (boto3 or aws_sdk_bedrock_runtime?)

I’m also currently trying to evaluate Nova Sonic using Python, and I’m running into issues installing dependencies.

I keep getting errors when trying to install aws_sdk_bedrock_runtime — specifically:

ERROR: No matching distribution found for aws_sdk_bedrock_runtime

Because of this, I’m not sure whether I should be using boto3 or the aws_sdk_bedrock_runtime package (which I can’t seem to install at all). I noticed you’re using Python — may I ask: are you using boto3, or have you somehow managed to get aws_sdk_bedrock_runtime working?

and if you used aws_sdk_bedrock_runtime can help me how. can I use it as package the installation in this example its self : https://github.com/aws-samples/amazon-nova-samples

Any tips would be greatly appreciated! Thanks!