Re: Classifying 'tone' of voice during speech recognition
Thanks for your response. I agree with you. I would love to be famous and
might just do that ;-). This is not my domain at present and I have other
things to worry about in addition to paying my bills with the current job. I
like to work on crazy challenging problem and this seems just the kind of
trouble I am usually looking for.
On Mon, Apr 5, 2010 at 3:22 PM, Jeff Brower wrote:
> > I am a newbie to speech recognition and DSP field. I am looking
> > for a software library that will take an audio file
> > and classify the 'tone' of voice in audio file to one of the few
> > predetermined categories e.g. sad, angry, annoyed,
> > happy, unknown, 'insufficient data' etc. In case of audio file
> > having multiple tones I would like to know ranking e.g.
> > 1-Angry (50%)
> > 2-Annoyed (40%)
> > 3-Sad (5%)
> > 4- Unknown (5%)
> > Can someone direct me to right resources.
> This is an area of active research... current methods appear to be
> "promising", not reliable. If you can create such
> a library that is content-accurate to 5%, you will be famous.
It is not because things are difficult that we do not dare; it is because
we do not dare that they are difficult.