Test SSML Examples with the Audio Sandbox

Some Speech Synthesis Markup Language (SSML) examples in the documentation display in the audio sandbox. You can use this sandbox to listen to the audio generated by the SSML and experiment with SSML tags.

Listen to SSML examples in the sandbox

The audio sandbox requires a developer account to generate and play the audio. If you have an Amazon retail account, you must complete developer account registration before you can generate audio in the sandbox. For details about creating a new developer account or linking an Amazon.com account to a developer account, see Create Your Amazon Developer Account.

To listen to an example in the sandbox

  1. If you aren't signed in to the developer portal, click Sign in and follow the prompts to sign in with your developer account.

  2. Click Load Audio. This action generates the audio and displays a waveform that represents the sound. 3/ Click Play to hear the audio.

    To start playback from a different position or jump to a different position during playback, click in the time line.

If you make changes to the code, click Reload audio to generate a new clip that reflects your changes.

Copy the example SSML code into your skill

The SSML examples shown in the sandbox wrap all SSML in <speak> tags and include line breaks for readability. When you copy the code to a JSON response in your skill, remove the line breaks.

If you use the ASK SDK, the SDK wraps outputSpeech with <speak> tags automatically. Remove these tags from the sample when you copy the code to a skill.

Sandbox limitations

The code editor doesn't save your changes. When you refresh the page or click the Reset button, the code shown in the editor reverts to its original state. Use the Copy button to copy your code to save or use within a skill.

The sandbox behaves as an English US device. SSML examples play using the default US English voice unless you use the voice and lang SSML tags.

Edits you make in the sandbox aren't saved. Use the Copy button to copy the code if you want to keep it.

Known issues

  • The emotion and domain tags aren't supported.
  • Special characters don't play correctly, for example, the characters used in phonemes.
  • If you sign in with an Amazon.com account that isn't linked to a developer account, attempting to load the audio displays the error "We had an issue with the request. Please come back and try later." To solve this issue, complete the developer account registration. For details, see Create Your Amazon Developer Account.
Was this page helpful?

Last updated: Nov 28, 2023