Thread: Speech, text, etc.

  1. #16
    Registered User Xei's Avatar
    Join Date
    May 2002
    Posts
    719

    Maybe

    Maybe, however how many programmers in the world have a need for voice recognition? I'm sure that if you had a goal set to do it then you would figure it out. Then along the way make updates, since there are so many ways to sample every frequency & frequency change to figure out what you are saying then there is not just one way to do it. You could use plenty of different algorithms that are equally comparable. But in the end it is just comparisons to frequencys and their changes in order to estimate what word the user is attempting to say. So once you can gather sound into its frequency's then do the comparisons I'd say your off to a real good start.

  2. #17
    ....
    Join Date
    Aug 2001
    Location
    Groningen (NL)
    Posts
    2,380
    >Maybe, however how many programmers in the world have a
    >need for voice recognition?

    A very little part of all programmers are working in the field of speech processing.

    >So once you can gather sound into its frequency's then do the
    >comparisons I'd say your off to a real good start.

    From a mathematical/technical point of view it's not hard to get the frequency spectrum of a signal and let some filters work on it. But it are the filters which are the hard part of speech processing, no voice is equal, so you would need adaptive filters and a good knowledge about human speech.

    The same is with audio coding. Compression algorithms for audio and for data are not hard to implement. But it takes a lot of testing to get a good quality audio coder.

  3. #18
    Registered User Xei's Avatar
    Join Date
    May 2002
    Posts
    719

    Yes

    Yes, getting an adaptive filter to work successfully would take some practice and time. I think how it would have to be done is to take an average of what frequency change each character said has; for example: if you say "Hello" the average person that speaks english should have approximately the same frequency change between the H and the e regardless of the pitch of their voice. At least thats how I think it would be done, then ofcourse there may be some software-like troublshooting to try to find out how fast they speak etc... and how long each character the person speaks is.

  4. #19
    Registered User Dual-Catfish's Avatar
    Join Date
    Sep 2001
    Posts
    802
    I'll be able to do Voice Recognition one day. I really will.
    That's great, I'm so proud of you.. but right now you fall into the 99% category who cannot program this from scratch.

    And about that 99% of the programmers cant do it? Hah! My uncle did before 1995.
    Wow, I'm so proud of your uncle. But what relevance does this have to the assumption that 99% of all programmers cannot do it?

    Ofcourse you can do it, just put your mind to it, learn what it involves, learn the mathematics, and ofcourse the API for the soundcard(no, not Win32 API).
    Mmm, this reminds me of all the newbies who come and want to program an OS, or a game which ranks up with StarCraft or Quake 3. Sure, if I decided today that I wanted to program a text-to-speech program, i'd eventually finish... but until then; guess what? I side with the 99% who cannot do it.

  5. #20
    Registered User Prodigy's Avatar
    Join Date
    May 2002
    Posts
    25

    ya

    sure why not
    And the cows moo...
    AIM:ProdigyCpp
    E-Mail:[email protected]
    If you like my avatar plz dont copy it go to:
    http://www.dragid.com/

  6. #21
    Registered User Xei's Avatar
    Join Date
    May 2002
    Posts
    719

    lol,funny man: Dual-Catfish

    lol, what relevance? Well if you read the board before making your own assumptions you might have re-phrased that.

    All of that message was in regards to the person saying "Could you do it? No." because that is an extremely negative attitude.

    You know what else makes me mad? When people say things like this:
    >"Mmm, this reminds me of all the newbies who come and want to program an OS"

    It reminds you of a newb eh? Good for you I'm so proud of you too!! But what does your comment have to do with relevance? My first message was just aimed at giving him confidence then the rest of my posts were just general conversation. But if shooting down other peoples posts gives you that warm feeling inside... just go ahead and continue.

  7. #22
    Registered User Dual-Catfish's Avatar
    Join Date
    Sep 2001
    Posts
    802
    You're new around here aren't ya Xei? Stick around for awhile, I think you'll be reading many posts like this one. I'm not going to get into a flame war with you, because that's exactly where this is headed.

    Give him all the confidence in the world... but I'll just prepare him for when he returns to reality.

  8. #23
    Registered User Xei's Avatar
    Join Date
    May 2002
    Posts
    719

    Thumbs up Even Funnier! lol

    Yes I am new. I am not trying to be mean, but I dont think you read the post very well. But oh well, whatever, maybe we both just don't understand eachother or something . Reality doesnt say that he can't do it, and that is what my whole post was about.

    -Confidence breeds success.

  9. #24
    Registered User
    Join Date
    Dec 2001
    Posts
    206
    a good idea is to learn the way .wav files are written and check to see if the default file is within 50 ascii codes of the ones the... something usmthin, i dunno, delete the post if u think its spam lol

  10. #25
    Registered User
    Join Date
    Apr 2002
    Posts
    139
    I wrote a text to speech program just the other day here is a section of the code
    Code:
    cout << "Give me your word";
    cin >> text_string;
    
    if (text_string=="lostminds")
      Play_wav("lostminds.wav");
    "The most common form of insanity is a combination of disordered passions and disordered intellect with gradations and variations almost infinite."

  11. #26
    Registered User
    Join Date
    Dec 2001
    Posts
    206
    whats the lib for Play_wav?

  12. #27
    Registered User
    Join Date
    Apr 2002
    Posts
    139
    Tis from the lostminds library
    "The most common form of insanity is a combination of disordered passions and disordered intellect with gradations and variations almost infinite."

Popular pages Recent additions subscribe to a feed

Similar Threads

  1. My text doesn't display
    By joeprogrammer in forum Game Programming
    Replies: 11
    Last Post: 02-23-2006, 10:01 PM
  2. Appending text to an edit control
    By dit6a9 in forum Windows Programming
    Replies: 3
    Last Post: 08-13-2004, 09:52 PM
  3. Text positioning and Text scrolling
    By RealityFusion in forum C++ Programming
    Replies: 3
    Last Post: 08-13-2004, 12:35 AM
  4. Scrolling The Text
    By GaPe in forum C Programming
    Replies: 3
    Last Post: 07-14-2002, 04:33 PM
  5. Replies: 1
    Last Post: 07-13-2002, 05:45 PM