formant analysis with the Java Speech API?

[otf]

I don't even know how good the Java Sound API is for grabbing a data feed
fielded by a mic and playing it on speakers as it was produced by a
speaker/singer.

Does anyone know of actual formant analysis and speech recognition with the
the Java Speech API?

Which issues would factor in?

How could you, for example, take into account that great many people speak
English with an accent or very deviant local versions? ;-)

The reason why I need to use Java is because the app should/will be a web
based one with no installation requirements and such

Any good links, sites, books on the subject?

otf

 

[Oliver Wong]

I don't even know how good the Java Sound API is for grabbing a data feed
fielded by a mic and playing it on speakers as it was produced by a
speaker/singer.

Does anyone know of actual formant analysis and speech recognition with
the
the Java Speech API?

Which issues would factor in?

How could you, for example, take into account that great many people speak
English with an accent or very deviant local versions? ;-)

The reason why I need to use Java is because the app should/will be a web
based one with no installation requirements and such

Any good links, sites, books on the subject?

There is, of course, the Java's Speech API Programmer's guide
(http://java.sun.com/products/java-media/speech/forDevelopers/jsapi-guide/index.html),
but from what I can see there, there is no direct support for formant
analysis. In fact, I believe the term "Formant" only appears once in that
entire document, in Chapter 2, and only in passing: "Most current systems do
it in one of two ways: concatenation of chunks of recorded human speech, or
formant synthesis using signal processing techniques based on knowledge of
how phonemes sound and how prosody affects those phonemes."

- Oliver

 

[onetitfemme]

There is, of course, the Java's Speech API Programmer's guide
(http://java.sun.com/products/java-media/speech/forDevelopers/jsapi-guide/index.html),
but from what I can see there, there is no direct support for formant
analysis. In fact, I believe the term "Formant" only appears once in that
entire document, in Chapter 2, and only in passing: "Most current systems do
it in one of two ways: concatenation of chunks of recorded human speech, or
formant synthesis using signal processing techniques based on knowledge of
how phonemes sound and how prosody affects those phonemes."

- Oliver

Well, apparently these engineers at sun microsystems are not
interested in audio/accoustical DSP and these kinds of things
..
and by the way. The smiley when I previously said things about people
having an accent was meant as a technical joke, meaning how hard things
would then get
..
otf