Publications

1. Encoder-Decoder Based Architecture for Fast and Real Time Audio Style Transfer

Karmakar S. , Ahmed A. , Panda A., Tangri P. and Ramani D.

To appear at Mediterranean Conference on Pattern Recognition and Artificial Intelligence (MedPRAI), 2019 at Istanbul, Turkey. Proposed a novel encoder-decoder based architecture for style transfer in audio. In this paper we created a single pass network for stylizing audio.

https://arxiv.org/abs/1812.07159

2. VFNet: A Convolutional Architecture for Accent Classification

Ahmed A. ,Tangri P. ,Panda A. , Ramani D. and Karmakar S.

To appear at IEEE INDICON 2019 at Rajkot, India. In this we worked on tackling accent classification. As accent is low level feature we used low level convolutional filters of variable sizes to extract the relavent feautures for further classificaiton. The results obtained were significantly better than previously done studies.

https://arxiv.org/abs/1910.06697

3. A Novel Architecture for Accent Transfer

Panda A. ,Tangri P. , Ahmed A. , Karmakar S. and Ramani D.

The publication is currently under review. In this we worked on Transfer of Accent using an encoder- decoder based architecture.