Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 28

Thread: What hardware / distro are you using ?

  1. #11
    FLAC TheLlama's Avatar
    Join Date
    Jul 2004
    Location
    All over the world
    Posts
    970
    Quote Originally Posted by intuitionsys
    [*]Festival (voice feedback)
    Sweet, I'm planning on using Festivus too. I wish it had some sort of built in compressor/expander so I could make the voice sound like a DJ.

    Quote Originally Posted by intuitionsys
    [*]Write my own FFT routines for the spectrum analyzer thus removing reliance on aRts and KDE
    I'm interested in collaborating with you on this one. I'm currently using FMOD because it provides me with visualization data. I would use straight-up ALSA (and probably mplayer) if I could just get Waveform and spectrum data.

    Oh Linux, you are so good to me.

  2. #12
    Maximum Bitrate
    Join Date
    Aug 2004
    Location
    at home
    Posts
    591
    Festival is interesting but there's also the Sphinx project which is really worth to spend time with it. Unfortunately for me it works only for English language, so i'm out of the competiton as teaching (training) a language to Sphinx requires sound studio with heavy hardware and a lot of time. btw it's out of reach from French users but it's really efficient anyway.
    It's strenght is far beyond the usual voice recognition systems. Have a look at it, compile it (it's very long), try it, it's an amazing tool that some U.S. army services aim to use for automated real time translation.
    Some experiments have also been made on operations theathers to communicate with local populations.

    About the fft
    Jack server may be usefull for what you want to do, alsa+fftw, there's a bunch of tools to help reach this target, at least you will find spectrum analyser and gauges sources to use with jack.
    First i thought about this software that this would add one more layer which would reduce the overall speed but it doesn't at all. It's fast and the results are pretty good with a small footprint

  3. #13
    FLAC TheLlama's Avatar
    Join Date
    Jul 2004
    Location
    All over the world
    Posts
    970
    A Festival/Sphinx combo would be really cool. I'm sure that is something I'll play with after I have some more essencial features of the player completed. Great, I will have to check out Jack. If it doesn't do what I want I suppose I can use FMOD.

  4. #14
    Variable Bitrate intuitionsys's Avatar
    Join Date
    Jul 2005
    Location
    Northern Virginia
    Posts
    293
    Quote Originally Posted by TheLlama
    I'm interested in collaborating with you on this one. I'm currently using FMOD because it provides me with visualization data. I would use straight-up ALSA (and probably mplayer) if I could just get Waveform and spectrum data.
    Once I get my studio gear all hooked back up and the finishing touches done, I'd like to get back to more Silverwolf development so yeah I'm game. I'll do some reading on FMOD to see what that's all about. The hard part as you're probably aware is not so much the FFT routines themselves but getting the data to them (and back out) without screwing something else up.

    Quote Originally Posted by kraft
    Festival is interesting but there's also the Sphinx project which is really worth to spend time with it. Unfortunately for me it works only for English language, so i'm out of the competiton as teaching (training) a language to Sphinx requires sound studio with heavy hardware and a lot of time. btw it's out of reach from French users but it's really efficient anyway.
    Well there are only a few voices I've found for Festival that sound really natural in any language. One of the things I'd like to do with Silverwolf too is have it bilingual (English and French). I've actually been thinking of using more pre-recorded stuff for static responses. I've already replaced some of them with a sexy female voice that are just wav files. Things like "MP3 Player", "Import Files", "Exiting", etc.

    Oh and don't get me started about heavy studio hardware - I almost dislocated my shoulder trying to wire up some stuff in a rack a week ago
    Silverwolf 2 is dead.
    Silverwolf 3...?

  5. #15
    Maximum Bitrate
    Join Date
    Aug 2004
    Location
    at home
    Posts
    591
    Quote Originally Posted by intuitionsys
    Well there are only a few voices I've found for Festival that sound really natural in any language. One of the things I'd like to do with Silverwolf too is have it bilingual (English and French).
    You would prefer add and use Mbrola project on top of festival, the voices are more natural and they are a lot of languages that can work with festival, have a look there

    Quote Originally Posted by intuitionsys
    I've actually been thinking of using more pre-recorded stuff for static responses. I've already replaced some of them with a sexy female voice that are just wav files. Things like "MP3 Player", "Import Files", "Exiting", etc.
    I agree it will give the best results and the most straight way but only for predictable events, the hassles comes when it's about having something in relation with the gps or whatever else which is subject to generate undefined informations then it becomes difficult to manage any case.

    Quote Originally Posted by intuitionsys
    Oh and don't get me started about heavy studio hardware - I almost dislocated my shoulder trying to wire up some stuff in a rack a week ago
    Lol i mispelled that i'm sorry, i wanted to say that serious sound hardware is required to train the Sphinx

  6. #16
    Variable Bitrate intuitionsys's Avatar
    Join Date
    Jul 2005
    Location
    Northern Virginia
    Posts
    293
    Quote Originally Posted by kraft
    You would prefer add and use Mbrola project on top of festival, the voices are more natural and they are a lot of languages that can work with festival, have a look there
    Yes Mbrola voices are better, although the only one I really took a liking to was the one English one with a british accent. AT&T rules for speech synthesis but the last I heard their development license was phenomenally large

    Quote Originally Posted by kraft
    Lol i mispelled that i'm sorry, i wanted to say that serious sound hardware is required to train the Sphinx
    I have alot of "serious" sound hardware in my studio but some of it is also "seriously heavy" I knew what you meant I just wanted to ***** about twisting my shoulder
    Silverwolf 2 is dead.
    Silverwolf 3...?

  7. #17
    Maximum Bitrate
    Join Date
    Aug 2004
    Location
    at home
    Posts
    591
    The French one from Mbrola needs some phonemes correction especially for the words with accents but they are near a natural pronunciation. I haven't the ear enough trained to appreciate as i should the English language

    So you hurt yourself with heavy hardware ? send it to /dev/null and give me the address where you empty your trash i will collect it

  8. #18
    FLAC TheLlama's Avatar
    Join Date
    Jul 2004
    Location
    All over the world
    Posts
    970
    Just to clarify, since i just found this out myself, Mbrola isn't actually a TTS. It is just a package of voices for festival

  9. #19
    Maximum Bitrate
    Join Date
    Aug 2004
    Location
    at home
    Posts
    591
    Mbrola is not a Text to Speech in it's standalone package, Mbrola is a voice synthesizer and mbrola provides enrichment for other languages than English but Festival + Mbrola works great

  10. #20
    Variable Bitrate intuitionsys's Avatar
    Join Date
    Jul 2005
    Location
    Northern Virginia
    Posts
    293
    Mbrola does work well with Festival but I just haven't bothered to use it (Mbrola) in my car pc. There were some Festival voices I found a while back I wanted to try but haven't gotten around to it.

    As an aside I've been looking at FMOD more seriously to embed music playback into Silverwolf instead of using the MPG123 backend. Thanks Llama for the pointer.
    Silverwolf 2 is dead.
    Silverwolf 3...?

Page 2 of 3 FirstFirst 123 LastLast

Similar Threads

  1. Replies: 43
    Last Post: 10-31-2006, 12:00 PM
  2. Distro block or no distro block?
    By JeremyM in forum Car Audio
    Replies: 15
    Last Post: 02-08-2006, 02:40 AM
  3. Another Linux distro for car PC?
    By COST Linux in forum Linux
    Replies: 12
    Last Post: 06-14-2005, 03:24 PM
  4. PyCar Linux distro TS check( EPIA-M )
    By jbors in forum Linux
    Replies: 17
    Last Post: 05-04-2005, 02:33 PM
  5. Possible EPIA distro for Linux, Can we make one?
    By adamis in forum Software & Software Development
    Replies: 15
    Last Post: 02-23-2005, 06:50 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •