Posted On: Mar 22, 2018
Amazon Polly is a Text-to-Speech service that uses advanced deep learning technologies to synthesize speech that sounds like a human voice. The new SSML Breath feature mimics the sound of inhalation and/or exhalation during normal speech. With the SSML Breath tag, you can now insert appropriate breath sounds to make the Amazon Polly generated speech sound even more natural, as if the text is being narrated by a human speaker.
You can use the <amazon:breath> and <amazon:auto-breaths> tags, or a combination of both, to incorporate breath sounds into your speech output.
- Automated mode: Amazon Polly inserts breathing sounds automatically into the speech output, to simulate normal breathing without input from you. (You can also set the frequency of these intervals, as well as their volume and duration.)
- Manual mode: you set the location, duration, and volume of a breath sound for the text.
- Mixed mode: a combination of customized breathing sounds that you insert, and that are inserted automatically by Amazon Polly.
By incorporating breath sounds into speech output from text, Amazon Polly is able to provide more naturally sounding speech, particularly for long-form narration of text.