Solving the natural speech recognition problem across languages & accents
The other day I wrote about the new voice search feature included on their site by Flipkart and CEans reported that the service didn't work 100% for them. I believe there's a big problem to be solved in the computer science domain and it has the potential to change the way people use computers and other computing devices - because that would make the next big shift in computing.
Right now, speech recognition works only with English language and that too mostly with American and British accents. Of course, this is changing rapidly; but it'll still be a long time before there's a unified approach / solution that works across various languages and accents.
At present, even English language recognition fails with different accents. Do you propose any solution to this problem? You'd be creating the next revolution in computing!
Right now, speech recognition works only with English language and that too mostly with American and British accents. Of course, this is changing rapidly; but it'll still be a long time before there's a unified approach / solution that works across various languages and accents.
At present, even English language recognition fails with different accents. Do you propose any solution to this problem? You'd be creating the next revolution in computing!
0