Custom pronunciations

The rule type is currently limited to voice clones, but we are working on making it available for our standard voices in the future.

Create a ipa_transcription, x_sampa_transcription, pinyin_transcription, or jyutping_transcription rule to force a word to be pronounced according to your preferences.

For example, you could create a rule that uses seˈβiʎa as the ipa_transcription for the word Seville.

You can apply this rule to all audios in a project or just a single audio.

NOTE: You should not add rules for numbers on their own unless their scope is for a single audio only. Issues with numbers should instead be reported to us through the Voice Issue tab.

Dashboard

• Use Key to enter the word you'd like to transcribe the pronunciation for.

• For Rule select the IPA, X-SAMPA, Pinyin, or Jyutping transcription rule.

• Use Value to enter the phonetic transcription.

API

Use the /rules endpoint to create new rules and the /content endpoint to regenerate to apply new rules to older content.