08-20-2016, 09:38 AM
Ooh, got a cool idea.
So I've been thinking about ways a mute character could communicate with people who don't know sign language and can't read lips.
I was thinking "type things out on phone text to speech" for one way, but then I realized...
Lip reading app. I'm honestly surprised such a thing doesn't appear to exist in reality.
Basically, using algorithms and crap (so specific) and the phone's front-facing camera, it'd read the lips of the mute person and text to speech what they're saying automatically, basically allowing them to speak as long as their phone's camera can see their lips. Granted, such a thing would probably be inaccurate, but I imagine it could be improved upon. The technology exists already, and has been used to read something Hitler said on video that couldn't be heard, apparently. There's like a documentary on it or something. And that's old, grainy video that was probably at an angle. It could totally be used with modern cameras. Also, unlike text to speech, since it would be in real-time, it could potentially gauge pauses better too.
So I've been thinking about ways a mute character could communicate with people who don't know sign language and can't read lips.
I was thinking "type things out on phone text to speech" for one way, but then I realized...
Lip reading app. I'm honestly surprised such a thing doesn't appear to exist in reality.
Basically, using algorithms and crap (so specific) and the phone's front-facing camera, it'd read the lips of the mute person and text to speech what they're saying automatically, basically allowing them to speak as long as their phone's camera can see their lips. Granted, such a thing would probably be inaccurate, but I imagine it could be improved upon. The technology exists already, and has been used to read something Hitler said on video that couldn't be heard, apparently. There's like a documentary on it or something. And that's old, grainy video that was probably at an angle. It could totally be used with modern cameras. Also, unlike text to speech, since it would be in real-time, it could potentially gauge pauses better too.