I have been using Amazon Polly to create custom sounds. I saved in OGG format and used sox to convert to sln, sampled at 32000 which I then renamed to sln32. The only file format I uploaded is sln32 which I assume are dynamically formatted to the client codecs like G729 as required. Is this the correct approach? Should I convert to some other formats as well?
I want to change the default core sounds so that I have the same voice. What’s the best way to go about this? Is there a listing of the default sounds and when they are called?