Just saw Apple dropped a new SpeechAnalyzer API. Interesting stuff. Looks like it’s designed for more detailed speech analysis, not just basic transcription.
My first thought: this could be pretty neat for Dreamers Dash. We’ve been kicking around ideas for audio-based clue verification, and this might be the ticket. Instead of just simple keyword spotting, maybe we could use this to understand more nuanced spoken responses or even verify if a player is mimicking a specific sound or phrase.
Definitely something to dig into. If it’s robust enough, it could make audio interactions in the game way more interesting than just tapping a button or typing.
Here’s the announcement video if you want to check it out: https://www.youtube.com/watch?v=0m6dimDDj8M
Need to add this to my list of things to experiment with. Could be a cool way to level up some of the puzzle mechanics in Dreamers Dash.
The SpeechTranscriber seems particularly interesting. It currently supports Chinese, English, French, German, Italian, Japanese, Korean, Portuguese, and Spanish. That’s most of the languages I will need for support in Epcot’s world showcase.