There are now dozens of AI-powered apps for turning text to speech, and the resulting speech is usually a monologue. When you start, the app will usually ask you to choose a voice. The other day, I started to wonder whether these apps could also create a dialogue consisting of 2 voices. If the answer was yes, that would be a great bonus to language teachers.
I then did an online search, and found a few apps which claimed to be able to do that. I then headed to Narakeet, which I had used before to convert a PPT with speaker notes to a narrated video. It turned out it was actually quite easy. After pasting in your dialogue text, just insert some 'stage directions' indicating which voice (among those provided) to use for which segment of the dialogue. Below is a TTS conversation which I have generated from a dialogue in Ch 1 of Open English I (an open-source textbook project). (Yes, it's very short, because you have to pay in order to get a longer dialogue audio.)
This edtech development is good news to language teachers, but bad news to those people who work as voice artists.
No comments:
Post a Comment