Strength is Life and Weakness is Death.
- Swami Vivekananda
Hey, I am Nirmesh J. Shah Ph.D. Student working under guidance of Prof. (Dr.) Hemant A. Patil at Speech Research Lab, DA-IICT, Gandhinagar, India. Currently, I am working in the area of Voice Conversion. I have started to explore Generative Adversarial Network (GAN) for Voice Conversion. Voice Conversion is the supervised learning problem. However, obtaining aligned speech pairs from both the source and target speakers' is itself very challenging. We identified potential issues related to the different alignment techniques in the context of different VC tasks and proposed the GAN-based architecture which avoids the need of alignment. In particular, we propose to use unsupervised Vocal tract Length (VTL) warped posterior features in GAN-based VC framework. I thank all my Speech Research Lab colleagues for their support. In particular, I would like to thank my collaborators, Neil Shah, Dr. Maulik C. Madhavi, Hardik B. Sailor, Sushant V. Rao, Sreeraj R., Avni Rajpal, Pramod Bachhav and Mohammadi Zaki.
During my master's at DA-IICT, Gandhinagar (2011-2013), I worked on development of HMM-based Speech Synthesis System (HTS) for Gujarati Language under the guidance of Prof. (Dr.) Hemant A. Patil. I worked as Research Assistant in Department of Electronics and Information Technology (DeitY), Government of India Sponsored project, "Development of Text-to-Speech Synthesis System for Indian Languages Phase-II" at DA-IICT, Gandhinagar (July 2012- December, 2015).
I love to discuss spiritual concepts (but very bad at following them :p ), plays harmonium, flutes and learning Ukulele. I can speak Gujarati, Hindi, English and currently learning Spanish (
Español) and Marathi.
Whenever I Feel Low : My source of motivation:
1) Imdad Ali Nandoliya (My Teacher who is very much dear to me.): Remembering him and his teachings immediately bring immense pleasure to me.
2) Swami Vivekananda: Complete Works
3) Swami Sivananda: Serve, Love, Give, Purify, Meditate and Realize.
Voice Conversion, Speech Synthesis, Speech Recognition, Speech Signal Processing, Machine Learning, Deep Learning and Statistics.
CEP 006, Speech Research Lab,
Gujarat, India- 382007.