algorithm - How to go about making an untrained speech to text converter? -
I am serious for severe deafness from a very young age but fortunately I can talk like a normal person. Oral communication has always been difficult for me because of my poor voice capabilities along with lip-reading, I have just gone through school and college through reading boards, powerpoint slides, books and internet. I used to present my current software engineering I am doing very well at the job, but late I feel that I should make some effort to improve my position.
Subtitles are my lifeguards in this country who understand movies / shows TV and I am enjoying it for the last 7 years (I am 31 years old).
I strongly feel the need to see the subtitles in real life whenever I talk to a person, even strangers. I want to develop an untrained speech for the text converter, and as a start, I do not even have to type the exact words, the signal only on syllables / phoneticity will also be cured.
I have been for a while, but most of the results either read text or speech recognition for the speech to make half-baked efforts on the computer to give voice commands. I have really got some indication on how to start this project. In particular, I need to do such steps as to how to handle audio files and to process almost everything to get the phoneticity as soon as possible.
You can ask to see the speech in real time. Some of them try it out.
Comments
Post a Comment