I generate my own mp3-files using Amazon Polly, which are outputted using the AudioPlayer interface. I tried different settings with ssml. Without it is too quiet. The last thing I tried was
<speak><prosody volume="x-loud"> <amazon:effect
to generate the audiofile,
but it is still quiter than the Alexa voice or the polly-voice using ssml (https://developer.amazon.com/en-US/docs/alexa/custom-skills/guidelines-ux-amazon-polly-skills.html: <voice name="Salli">TEXT</voice> ). Are there any advice how to achieve the best result? I want to imitate <voice name="Salli">TEXT</voice> as much as possible.
In case there are differences for the best settings depending on the Polly voice, I would be interested in: Salli (EN-US: neural ) , Joanna(EN-US: neural), Marlene (DE), Vicky (DE), Amy(EN-UK: neural), Emma (EN-UK:neural).