Automatic glottis localization and segmentation in stroboscopic videos
About
In speech production, the vocal folds play a crucial role in modulating airflow from the lungs through their quasiperiodic vibrations. Between the vocal folds lies a narrow opening, known as the glottis, which allows air to pass through the trachea. Changes in the shape or muscular properties of the vocal folds alter the glottal opening, leading to variations in voice quality, such as hoarseness and dysphonia. Many of these conditions result in an incomplete closure of the glottis during voice production, a phenomenon referred to as a glottic chink. Speech-language pathologists (SLPs) typically visualize the glottic chink using endoscopic or stroboscopic video. This project aims to automate the localization and segmentation of the glottis in stroboscopic videos.>>top<<
Publications
Journals (Accepted and/or Published):
- Varun Belagali, Achuth Rao M V, Pebbili Gopikishore, Rahul Krishnamurthy, Prasanta Kumar Ghosh, "Two step convolutional neural network for automatic glottis localization and segmentation in stroboscopic videos", accepted in Biomed. Opt. Express 11, 4695-4713 2020. [PDF]
Conferences (Accepted and/or Published):
- Divya Degala, Achuth Rao M V, Rahul Krishnamurthy, Pebbili Gopikishore, Veeramani Priyadharshini, Prakash T K, Prasanta Kumar Ghosh, "Automatic Glottis Detection and Segmentation in Stroboscopic videos using Convolutional Networks", accepted in Interspeech 2020, Shanghai, China. [PDF] [Slides] [Presentation]
- Achuth Rao M V, Rahul Krishnamurthy, Pebbili Gopikishore, Veeramani Priyadarshini, Prasanta Kumar Ghosh, "Automatic glottis localization and segmentation in stroboscopic videos using deep neural network", accepted in In Proc. Interspeech, Hyderabad, India,2018, Page(s): 3007-3011. [PDF]
Rahul Krishnamurthy
