https://elevenlabs.io/docs/speech-synthesis/prompting How to use phoneme and break tags in 11labs, it seemed that all xml tags that LLM output are filtered out

@codeboy3516 It seems like you're looking to understand how to use phoneme and break tags with ElevenLabs' speech synthesis, but you're concerned that XML tags might be filtered out from the output. To use phoneme and break tags in ElevenLabs, you would typically include these tags within the SSML (Speech Synthesis Markup Language) that you send to their API. Here's a brief explanation: - **Phoneme Tag**: This tag allows you to specify the pronunciation of a word using phonetic alphabet or sounds. For example:

xml
  <phoneme alphabet="ipa" ph="təˈmeɪtoʊ">tomato</phoneme>

This will ensure that "tomato" is pronounced as specified by the phonetic alphabet. - **Break Tag**: This tag lets you insert pauses of different lengths within the speech. For example:

xml
  <break time="500ms"/>

This will create a half-second pause in the speech. If you're experiencing issues with XML tags being filtered out, it could be due to the way the LLM processes the output. To address this, you might need to escape the XML tags or use a CDATA section to ensure they are not stripped out. However, for specific instructions and support regarding ElevenLabs' API and how it handles SSML, you should refer to their documentation or contact their support team. If you need further assistance with this or have any other questions, please let me know!

enable ssmlParsing on your assistant and then you can use it.