Zobrazeno 1 - 10
of 37
pro vyhledávání: '"Agrawal, Purvi"'
Publikováno v:
2022 IEEE Spoken Language Technology Workshop (SLT), Doha, Qatar, 2023, pp. 252-259
Even with several advancements in multilingual modeling, it is challenging to recognize multiple languages using a single neural model, without knowing the input language and most multilingual models assume the availability of the input language. In
Externí odkaz:
http://arxiv.org/abs/2401.11645
In this work, we propose a multi-head relevance weighting framework to learn audio representations from raw waveforms. The audio waveform, split into windows of short duration, are processed with a 1-D convolutional layer of cosine modulated Gaussian
Externí odkaz:
http://arxiv.org/abs/2107.14793
Autor:
Agrawal, Purvi, Ganapathy, Sriram
Publikováno v:
IEEE International Conference on Acoustics, Speech, & Signal Processing (ICASSP) 2021
In this work, we propose an acoustic embedding based approach for representation learning in speech recognition. The proposed approach involves two stages comprising of acoustic filterbank learning from raw waveform, followed by modulation filterbank
Externí odkaz:
http://arxiv.org/abs/2102.07390
Autor:
Agrawal, Purvi, Ganapathy, Sriram
Publikováno v:
IEEE Transactions and Audio, Speech and Language Processing, Vol. 28, pp. 2823 - 2836, 2020
The learning of interpretable representations from raw data presents significant challenges for time series data like speech. In this work, we propose a relevance weighting scheme that allows the interpretation of the speech representations during th
Externí odkaz:
http://arxiv.org/abs/2011.02136
Autor:
Agrawal, Purvi, Ganapathy, Sriram
Publikováno v:
Proc. Interspeech 2020, 1649-1653 (2020)
Speech recognition in noisy and channel distorted scenarios is often challenging as the current acoustic modeling schemes are not adaptive to the changes in the signal distribution in the presence of noise. In this work, we develop a novel acoustic m
Externí odkaz:
http://arxiv.org/abs/2011.00721
Autor:
Agrawal, Purvi, Ganapathy, Sriram
Speech recognition from raw waveform involves learning the spectral decomposition of the signal in the first layer of the neural acoustic model using a convolution layer. In this work, we propose a raw waveform convolutional filter learning approach
Externí odkaz:
http://arxiv.org/abs/2001.07067
Autor:
Authreya, Ashwini J1 (AUTHOR) dr.ashwini2009@gmail.com, Agrawal, Purvi1 (AUTHOR), Makam, Adinarayana1 (AUTHOR)
Publikováno v:
Australasian Journal of Ultrasound in Medicine. May2021, Vol. 24 Issue 2, p70-77. 8p.
Publikováno v:
Journal of Fetal Medicine; Dec2022, Vol. 9 Issue 3/4, p105-108, 4p
Autor:
Agarwal, Riya, Soni, Purva, Agrawal, Purvi, Rathor, Sejal, Sisodiya, Sandhya, Sharma, Jitendra
Publikováno v:
International Journal of Advanced Research in Computer Science; 2022 Special Issue, Vol. 13, p125-129, 5p
Publikováno v:
Journal of Fetal Medicine; 6/1/2022, Vol. 9 Issue 2, p29-33, 5p