Zobrazeno 1 - 10
of 1 316
pro vyhledávání: '"Sarangarajan A"'
Autor:
Levit, Michael, Parthasarathy, Sarangarajan, Aksoylar, Cem, Rasooli, Mohammad Sadegh, Chang, Shuangyu
We propose an adaptation method for factorized neural transducers (FNT) with external language models. We demonstrate that both neural and n-gram external LMs add significantly more value when linearly interpolated with predictor output compared to s
Externí odkaz:
http://arxiv.org/abs/2305.17304
Autor:
Kuan-Wei Peng, Allison Klotz, Arcan Guven, Unnati Kapadnis, Shobha Ravipaty, Vladimir Tolstikov, Vijetha Vemulapalli, Leonardo O. Rodrigues, Hongyan Li, Mark D. Kellogg, Farah Kausar, Linda Rees, Rangaprasad Sarangarajan, Birgitt Schüle, William Langston, Paula Narain, Niven R. Narain, Michael A. Kiebish
Publikováno v:
Scientific Reports, Vol 14, Iss 1, Pp 1-8 (2024)
Abstract Parkinson’s disease is a progressive neurodegenerative disorder in which loss of dopaminergic neurons in the substantia nigra results in a clinically heterogeneous group with variable motor and non-motor symptoms with a degree of misdiagno
Externí odkaz:
https://doaj.org/article/7ef1ad93078f4891b7305fd1f6d91e67
It is challenging to train and deploy Transformer LMs for hybrid speech recognition 2nd pass re-ranking in low-resource languages due to (1) data scarcity in low-resource languages, (2) expensive computing costs for training and refreshing 100+ monol
Externí odkaz:
http://arxiv.org/abs/2209.04041
Autor:
Hiyama, Eiso, Hishiki, Tomoro, Yoshimura, Kenichi, Krailo, Mark, Maibach, Rudolf, Haeberle, Beate, Rangaswami, Arun, Lopez-Terrada, Dolores, Malogolowkin, Marcio H., Ansari, Marc, Alaggio, Rita, O’Neill, Allison F., Trobaugh-Lotrario, Angela D., Watanabe, Kenichiro, Schmid, Irene, Ranganathan, Sarangarajan, Tanaka, Yukichi, Inoue, Takeshi, Piao, Jin, Lin, Jason, Czauderna, Piotr, Meyers, Rebecka L., Aronson, Daniel C.
Publikováno v:
In eClinicalMedicine October 2024 76
Autor:
Sarangarajan, A.V., Jain, Adarsh, Ferreir, Jenifer L., Anushree, Dhanawat, Aniket, Ahir, Pankita, Acharya, Sanjeev
Publikováno v:
In PharmaNutrition September 2024 29
In recent years, end-to-end (E2E) based automatic speech recognition (ASR) systems have achieved great success due to their simplicity and promising performance. Neural Transducer based models are increasingly popular in streaming E2E based ASR syste
Externí odkaz:
http://arxiv.org/abs/2110.01500
Autor:
Meng, Zhong, Kanda, Naoyuki, Gaur, Yashesh, Parthasarathy, Sarangarajan, Sun, Eric, Lu, Liang, Chen, Xie, Li, Jinyu, Gong, Yifan
Publikováno v:
2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toronto, Canada
The efficacy of external language model (LM) integration with existing end-to-end (E2E) automatic speech recognition (ASR) systems can be improved significantly using the internal language model estimation (ILME) method. In this method, the internal
Externí odkaz:
http://arxiv.org/abs/2102.01380
Autor:
Ruach Sarangarajan, Cornelius Ewuoso
Publikováno v:
Frontiers in Public Health, Vol 12 (2024)
In this paper, we draw on the thinking about solidarity, reciprocity and distributive justice grounded in Afro-communitarian ethics from the Global South to argue for institutions, particularly the South African (SA) government, have a prima facie du
Externí odkaz:
https://doaj.org/article/7856e1e9bfbd4309963d4208f6adc1ad
Autor:
Meng, Zhong, Parthasarathy, Sarangarajan, Sun, Eric, Gaur, Yashesh, Kanda, Naoyuki, Lu, Liang, Chen, Xie, Zhao, Rui, Li, Jinyu, Gong, Yifan
Publikováno v:
2021 IEEE Spoken Language Technology Workshop (SLT)
The external language models (LM) integration remains a challenging task for end-to-end (E2E) automatic speech recognition (ASR) which has no clear division between acoustic and language models. In this work, we propose an internal LM estimation (ILM
Externí odkaz:
http://arxiv.org/abs/2011.01991
LSTM language models (LSTM-LMs) have been proven to be powerful and yielded significant performance improvements over count based n-gram LMs in modern speech recognition systems. Due to its infinite history states and computational load, most previou
Externí odkaz:
http://arxiv.org/abs/2010.11349