You can improve the rate of words recognised by our AI by creating a dictionary containing the pronunciation of difficult words.
To create a new dictionary or edit an existing one, go to the Vocabularies tab via ACCOUNT . All your Vocabularies are listed here. To add to an existing dictionary, click on "Edit" to the right of the Vocabulary. If you want to create a new Vocabulary, click on "Create a new vocabulary" at the bottom.
In the left column, enter the correct spelling of the word or phrase, which will then be displayed as the subtitle (this is the most important step). In the right column, you can optionally enter the pronunciation if there are any variations (e.g., “nyohki” for “gnocchi”). This can help improve recognition if necessary. You can also provide translations for your entries (learn more here).
The entries can now either be imported into the dictionary as a batch using an existing .csv or .txt file (learn more here) or added individually as entries. Once you’ve added all entries, click “Create Dictionary” (or “Update Dictionary” if you’re editing an existing one).
Please note: A maximum of 1,000 entries can be uploaded. We cannot process special characters such as &, °, %, or €.
To now use the Vocabulary in your SPEECH session, you have to select it when creating the session.
In an already created session, you can add a vocabulary in the "details". It is not possible to do that during audio-transfer, so you have to pause it for that.