Voice Diagnostics - Science topic

Dear Rajat Tandon

This article might be useful to know further have a look:

Kind Regards

Qamar Ul Islam

Hi,

Did you find solution for this..??

And also do you have any source to know about what are BAP features and how aperiodicity extracted from the signal..

Thanks

Giridhar

Hi Jeremy, 

Where were the electrodes placed? 

Was it telemetry? Could it be simply an ECG interference in the signal since the first peak with the highest amplitude is at around 5 Hz (rr interval mouse is 6-8 Hz)?

I noticed you used Neuroscore for the analysis. How often does it happen to have such a pattern?

Thanks.

It depends on the later processing.  If you project your features into some low dimensional space in a sensible way, then the additional delta features become redundant, and only the original features matter.  Check the IBM papers about a decade ago for more information.

For these kind of projects, once you recorded the person's voice, of course it will be saved as an audio file. Then you have to do feature extraction using Matlab Programming to remove the noises from the voice signals.

Besides, you have to train your system in advance. Meaning that you must build a database or use the available databases (perhaps u can find it online). By training the system, the system will learn how to detect the voices and in which category do these voices belong to. Then only once you give the voice as the "testing data" it will be able to detect the emotions.

There are lots of feature extraction and classification methods that u can use. Many articles available too.

Hope that gives u the general idea that how it works.

I just found this one online. It is a short article, but may give u more clear image of the story

Dear Yasir Rahmatallah ,Alessio Brutti ·,B. Tomas 

Thank you for the detailed answer. It is exactly what I was trying to do.

For the Voice Related Quality of Life (VRQOL) you should read article  by Hogikyan et al in Journal of Voice (2009) which describes the validation of the VRQOL and information is given on scoring. The raw scores are converted to standard scores using an alogorithm (in the article).

I do not have experience with the VOS.

Hope this helps

Patricia

Try the following ref:

[ ]s

Seek and apply for summer research opportunities where there is someone working in the field you are interested in. In the USA many Medical Students interested in research take a year off between their 3rd and 4th year of medical school to work in a mentored research arrangement.  I don't know what is available in your location but in the USA there are many Academic Institutions here that have funded programs available to facilitate this. 

I think, you have to detect both, voicing level of speech signal and unvoiced portions and the time depency of the transition.

Yes. You can measure nasality and there are signal proxessing techniques. In particular Nasality is an aspect of voice “quality” traditionally assessed by the perceptions of professionals involved in the evaluation and treatment of resonance disorders.  However, despite the importance of an expert’s perception of the absence, or presence and degree, of nasality to the evaluation process, there is also a need for objective measurement.

For this purpose clinical instruments like the Nasometer has provided an easy method for assessing nasality objectively.The nasality measure is derived from the ratio of acoustic energy output from the nasal and oral cavities of the speaker.  Using an innovative input device consisting of a directional microphone mounted on either side of an efficient sound separator plate (with 25 dB separation between channels), the Nasometer headset facilitates accurate signal analysis, which ultimately yields a “nasalance” score.  As the speech signal enters the system, the ratio of the nasal acoustic energy to the oral plus nasal acoustic energy is calculated in terms of percentage (i.e., (N/(N + O)) x 100 = Nasalance) and is displayed graphically on the host computer screen in real time.

You can find more details in the following link. Do not hesitate to contact me for any more details.

I hope this is useful.

Best Eugenia

What is the problem about Praat? It is scriptable, so if you know what you want to do, you can do it. Of course MDVP offers some nice routines. In case you want to work on the signal itself also Matlab might be an option (toolboxes for signal analysis of especially: http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html)

Hi ,

Stroboscopy will give the information on the mucosal wave