Vagmi Voice Diagnostics

Voice diagnostics is also referred to as acoustic analysis of voice or objective assessment of voice.

This software is useful to professionals such as:

  • Speech Language Pathologists
  • ENT Doctors
  • Otolaryngologists
  • Phono-surgeons
  • Teachers and Researchers interested in a study of voice

Goal of voice analysis

Ideally by (acoustic) voice analysis one must be able to:

  •  To identify objectively the underlying functional inadequacies or deficiencies present even in a normal voice
  • To identify the underlying type of disorder when there is a noticeable change in voice
  • Suggest ways to improve the voice by therapy, if possible
  • Monitor the changes in voice attributes and modify therapy strategy if required
  • To identify the factors that make a voice sound great or appealing to a listener for improving the voice quality of professionals using voice
  • Suggest if surgical intervention is required. If so, to monitor objectively the pre and post op voice characteristics

Broad classification of voice disorders

Functional Disorders
Voice is normal but mechanism is not used optimally
Voice abuse present
Restricted range of F0 and intensity
Voice is tired or hoarse or unclear after prolonged use
Not appropriate to gender and age
Etc.

Neurological Disorders
Poor control – too many fluctuations in F0 and intensity
Un-intended breaks or sudden changes in voice quality
Etc.

Organic Disorders
Edema, Nodules
Bowing
Paralysis
Etc.

Acoustic Analysis of Voice

In Vagmi, there are two versions:
Advanced Plus Version
Comprehensive Version

Vagmi Advanced Plus Version

In Vagmi Advanced plus version, the measurements relating to voice are broadly categorized in to the following groups:

Recording and registration
Facility to record the client’s voice sample for desired duration
Assessing the quality of recording
Saving the entire voice sample or part of it
Registration of the client and associating the voice sample file to the client

Respiratory measurements (On-line module)
Voice Initiation and Sustenance
Maximum Phonation Duration

Phonatory measurements
F0 and Intensity measure – Mean, Standard Deviation, Fluctuations
Jitter and Shimmer measurements – Cycle to cycle variations in speech/source
Harmonic to Noise Ratio (HNR)
Dysphonia Index (DSI)
Long term average spectrum (LTAS) of voice source; alpha, beta, gamma
Long term average spectrum (LTAS) of speech; alpha, beta, gamma

Other Measurements (On-line module)
Voice Screening
Maximum Phonation Duration (MPD)
S/Z Ratio
Nasalence (Requires customized hardware developed by VSS)
Diadacho kinetic rate (DDK)

Report Generation and Comparison with the Normative

Vagmi Comprehensive Version

Vagmi comprehensive version includes in addition to the above Vagmi Advanced version and the following measurements:

Laryngeal or Acoustic Glottography (Acoustic Glottogram)
Inverse filtering – LP and formant based techniques
Temporal measures: Opening, Closing, Closed, Leakage intervals relative to pitch period
Spectral measures: Harmonic ratio, Spectral Roll-off

Spectrographic Measurements
Wide-band, Narrow-band, very-wide-band options
Gray scale/coloured
Contrast, brightness control
Frequency axis – Hz, mel, Bark
Frequency range – full, one-half, one-quarter
Superpose measured F0

Vowel Synthesis with Perturbations
Vowels can be synthesized with controlled standard deviation, random perturbations, modulation (depth and frequency) of any one or combination of the voice parameters. Such a facility may be used to train a person to recognize perceptually the changes in voice. Also, the synthesized vowels may be used for validating the software. Synthesized vowels may be saved to conduct a perceptual test.

Documentation

The following manuals are supplied along with the software:
User’s Guide
Tutorial Manual
On-line Measurements Manual
Case studies Manual
Validation Manual
Quick Reference Guide
Technical Support Manual