Jump to content

Sign Language Translator

Hey guys,

I'm looking for something that can translate sign language in real-time. If not someone with the know how to help build and code something like it. My Girlfriends sister is Deaf and finds it hard to get a job and to interact with people so I was wondering how hard this actually could be to build. All you would need is some finger sensors on a glove and maybe a Rasberry Pi to learn the gestures. I mean phones learn simple gestures so why can't we programme it to learn ASL or BSL to translate and a speaker the deaf person can wear. Is this possible to build? It cant be that expensive either, can it? 

Any Responses are welcome, thank you. 

Link to comment
Share on other sites

Link to post
Share on other sites

It's a little tricky I'd assume, since sign language also has its dialects. This looks up your alley though:

https://www.newscientist.com/article/2133451-automatic-sign-language-translators-turn-signing-into-text/

http://www.kintrans.com

Crystal: CPU: i7 7700K | Motherboard: Asus ROG Strix Z270F | RAM: GSkill 16 GB@3200MHz | GPU: Nvidia GTX 1080 Ti FE | Case: Corsair Crystal 570X (black) | PSU: EVGA Supernova G2 1000W | Monitor: Asus VG248QE 24"

Laptop: Dell XPS 13 9370 | CPU: i5 10510U | RAM: 16 GB

Server: CPU: i5 4690k | RAM: 16 GB | Case: Corsair Graphite 760T White | Storage: 19 TB

Link to comment
Share on other sites

Link to post
Share on other sites

It was tried about a year ago: https://www.newscientist.com/article/2140592-glove-turns-sign-language-into-text-for-real-time-translation/

 

Ultimatly, it's going to be very difficult to make such a thing. While ASL movements are pretty standardized, there's still going to be subtly within each person.

There's also dialects and slang within ASL. 

 

Hell, look at how difficult it is for computers to recognize handwritting via a direct touch input. Just imagine how difficult it would be for a computer to translate ASL via a camera or sensors. 

Nova doctrina terribilis sit perdere

Audio format guides: Vinyl records | Cassette tapes

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, tikker said:

It's a little tricky I'd assume, since sign language also has its dialects. This looks up your alley though:

https://www.newscientist.com/article/2133451-automatic-sign-language-translators-turn-signing-into-text/

http://www.kintrans.com

Yeah, It'll be tricky but it's almost like every language and Google Translate has shown it's possible. I know Google has literally the biggest team behind any Linguistics software of it's kind but even if it's the trouble of such dialects businesses could still make money from selling each one separately. If you could use Googles API for Google Translate isn't there also a strong possibility that the actually translating to whatever language won't be hard because all you would need to do is decipher what the sign the person is doing is saying and simultaneously put it through Google Translator so it reads it out? 

Link to comment
Share on other sites

Link to post
Share on other sites

Surely with AI or Computer Learning software, it could detect the difference, or am I over simplifying this? 

Link to comment
Share on other sites

Link to post
Share on other sites

47 minutes ago, Njordic said:

Yeah, It'll be tricky but it's almost like every language and Google Translate has shown it's possible. I know Google has literally the biggest team behind any Linguistics software of it's kind but even if it's the trouble of such dialects businesses could still make money from selling each one separately. If you could use Googles API for Google Translate isn't there also a strong possibility that the actually translating to whatever language won't be hard because all you would need to do is decipher what the sign the person is doing is saying and simultaneously put it through Google Translator so it reads it out? 

The product I linked literally does that.

 

The thing with these kind of things is that it works best with a large team behind it if you go the machine learning route. It needs to be trained, have a large database etc. It's no coincidence that Google Translate is one of the (if not the) best around while simultaneously having a dedicated team and lots of resources.

Crystal: CPU: i7 7700K | Motherboard: Asus ROG Strix Z270F | RAM: GSkill 16 GB@3200MHz | GPU: Nvidia GTX 1080 Ti FE | Case: Corsair Crystal 570X (black) | PSU: EVGA Supernova G2 1000W | Monitor: Asus VG248QE 24"

Laptop: Dell XPS 13 9370 | CPU: i5 10510U | RAM: 16 GB

Server: CPU: i5 4690k | RAM: 16 GB | Case: Corsair Graphite 760T White | Storage: 19 TB

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, tikker said:

The product I linked literally does that.

 

The thing with these kind of things is that it works best with a large team behind it if you go the machine learning route. It needs to be trained, have a large database etc. It's no coincidence that Google Translate is one of the (if not the) best around while simultaneously having a dedicated team and lots of resources.

That's awesome tech but that uses a camera as the sensor so using it as a personal device is impossible, it's crazy to think no one's done this as a finger sensor. Thanks guys for the help :)

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, Njordic said:

That's awesome tech but that uses a camera as the sensor so using it as a personal device is impossible, it's crazy to think no one's done this as a finger sensor. Thanks guys for the help :)

Ah you were specifically looking for a glove type thing. Well then @Volbet's article looks to be the closest to that so far. It's not that nobody has done it, more that it may not be ready for production yet.

 

Here's a video of such a glove: https://www.youtube.com/watch?v=l01sdzJHCCM

Crystal: CPU: i7 7700K | Motherboard: Asus ROG Strix Z270F | RAM: GSkill 16 GB@3200MHz | GPU: Nvidia GTX 1080 Ti FE | Case: Corsair Crystal 570X (black) | PSU: EVGA Supernova G2 1000W | Monitor: Asus VG248QE 24"

Laptop: Dell XPS 13 9370 | CPU: i5 10510U | RAM: 16 GB

Server: CPU: i5 4690k | RAM: 16 GB | Case: Corsair Graphite 760T White | Storage: 19 TB

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×