About

One of the typical symptoms of neuro-degenerative diseases like Amyotrophic Lateral Sclerosis (ALS), Parkinson's Disease (PD) and Alzheimer's Disease is dysarthria, that is, difficulty in speech production. Dysarthria involves slurred/mumbled speech, too slow/fast speaking rate, very loud/soft speech, nasal, breathy and strained voice, long and frequent pauses. These impairments compromise the naturalness and often intelligibility of speech.

Objectives:

• To detect neuro-degenerative diseases automatically at early stages using speech cues
• To monitor the severity and progression of the diseases using speech cues
• To enhance the naturalness and intelligibility of dysarthric speech
• To understand and model the effects of these diseases on different parts of the speech production system
• To develop a mobile/web application that can assist neurologists in diagnosing the disease and monitoring its progression using speech as a biomarker

Background and Motivation:

The figure below shows speech waveforms and associated spectrograms for an ALS patient, a PD patient and a healthy individual. Each person is uttering the word ‘pa’ in succession. We can see that the speaking rate significantly reduces for the patients and the information content in the spectrogram gets confined to lower frequency regions. This project aims to identify and analyse such distinguishing patterns between dysarthric and healthy speech.

>>top<<

Funding

• Department of Science and Technology, Govt. of India

>>top<<

Publications

Journals (Accepted and/or Published):

0. Stay tuned....

Conferences (Accepted and/or Published):

1. Jhansi Mallela, Aravind Illa, Yamini Belur, Nalini Atchayaram, Ravi yadav, Pradeep Reddy, Dipanjan Gope, Prasanta Kumar Ghosh, "Raw speech waveform based classification of patients with ALS, Parkinson’s Disease and healthy controls using CNN-BLSTM", accepted in Interspeech 2020, Shanghai, China.
2. Suhas BN, Jhansi Mallela, Aravind Illa, Yamini BK, Nalini Atchayaram, Ravi Yadav, Dipanjan, Prasanta Kumar Ghosh, "Speech task based automatic classification of ALS and Parkinson’s Disease and their severity using log mel spectrograms", In 2020 International Conference on Signal Processing and Communications (SPCOM) (pp. 1-5). IEEE.
3. Jhansi Mallela, Aravind Illa, Suhas B N, Sathvik Udupa, Yamini Belur, Nalini Atchayaram, Ravi Yadav, Pradeep Reddy, Dipanjan Gope, Prasanta Kumar Ghosh, "Voice based classification of patients with Amyotrophic Lateral Sclerosis, Parkinson's Disease and healthy controls ith CNN-LSTM using transfer learning", In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, 2020.
4. Suhas BN, Deep Patel, Nithin Rao, Yamini Belur, Pradeep Reddy, Nalini Atchayaram, Ravi Yadav, Dipanjan Gope and Prasanta Kumar Ghosh, "Comparison of Speech Tasks and Recording Devices for Voice Based Automatic Classification of Healthy Subjects and Patients with Amyotrophic Lateral Sclerosis", Proc. Interspeech 2019, pp. 4564-4568, Graz, Austria.
5. Aravind Illa, Deep Patel, Yamini BK, Meera SS, Shivashankar N, Preethish Kumar Veeramani, Seena Vengalil, Kiran Polavarapu, Saraswati Nashi, Atchayaram Nalini and Prasanta Kumar Ghosh, "Comparison of speech tasks for automatic classification of patients with Amyotrophic Lateral Sclerosis and healthy subjects", In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018, pp. 6014-6018.

Posters (Accepted and/or Published):

6. Jhansi M, Aravind I, Yamini BK, Nalini A, Seena V, Saraswathi N, Ravi Y, Prasanta Kumar Ghosh, "Low frequency characteristics are the key differentiators between dysarthric speech in ALS and healthy speech", 2nd Asia Pacific Frontotemporal and Motor Neurone Disease Meeting (APFM) 2020. [Slide]

>>top<<