Amazon polly

Owned by Santiago Viglione
Last updated: Aug 03, 2022
1 min read

The activity provides the possibility to use the Text To Speech service of Amazon.

It invokes the service and generates the corresponding audio which is stored within “/tmp/” folder to reproduce it later with activities like Playback or Read.

When the audio is played using Playback or Read activities, you must specify the complete path: /tmp/${guid}

The configuration of the activity requires the next parameters

  1. Language https://docs.aws.amazon.com/polly/latest/dg/SupportedLanguage.html

  2. Voice (Must select a voice compatible with the selected language) https://docs.aws.amazon.com/polly/latest/dg/voicelist.html

  3. AccessKey of Amazon Polly

  4. SecretAccessKey of Amazon Polly

  5. Text (The text that has to be translated to voice).

Each time the application is called generates new audio with the call GUID as the name and overwrites the previous one.

For testing proposes, due to the negotiation time of RTP after the call is connected, it is recommended to use a delay in the answer activity with at least 1000ms to hear the audio correctly.