Výsledky vyhledávání - "Sekhar, C. Chandra"

Report

Memory-efficient Stochastic methods for Memory-based Transformers

Autor: Vishnu, Vishwajit Kumar, Sekhar, C. Chandra

Training Memory-based transformers can require a large amount of memory and can be quite inefficient. We propose a novel two-phase training mechanism and a novel regularization technique to improve the training efficiency of memory-based transformers

Externí odkaz: http://arxiv.org/abs/2311.08123

Zobrazit plný text záznamu

Report

Unsupervised Domain Adaptation in Speech Recognition using Phonetic Features

Autor: Ojha, Rupam, Sekhar, C Chandra

Automatic speech recognition is a difficult problem in pattern recognition because several sources of variability exist in the speech input like the channel variations, the input might be clean or noisy, the speakers may have different accent and var

Externí odkaz: http://arxiv.org/abs/2108.02850

Zobrazit plný text záznamu

Report

Front-end Diarization for Percussion Separation in Taniavartanam of Carnatic Music Concerts

Autor: Dawalatabad, Nauman, Sebastian, Jilt, Kuriakose, Jom, Sekhar, C. Chandra, Narayanan, Shrikanth, Murthy, Hema A.

Instrument separation in an ensemble is a challenging task. In this work, we address the problem of separating the percussive voices in the taniavartanam segments of Carnatic music. In taniavartanam, a number of percussive instruments play together o

Externí odkaz: http://arxiv.org/abs/2103.03215

Zobrazit plný text záznamu

Report

Novel Architectures for Unsupervised Information Bottleneck based Speaker Diarization of Meetings

Autor: Dawalatabad, Nauman, Madikeri, Srikanth, Sekhar, C. Chandra, Murthy, Hema A.

Publikováno v: IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, 2021, pp 14-27

Speaker diarization is an important problem that is topical, and is especially useful as a preprocessor for conversational speech related applications. The objective of this paper is two-fold: (i) segment initialization by uniformly distributing spea

Externí odkaz: http://arxiv.org/abs/2010.06304

Zobrazit plný text záznamu

Report

Incremental Transfer Learning in Two-pass Information Bottleneck based Speaker Diarization System for Meetings

Autor: Dawalatabad, Nauman, Madikeri, Srikanth, Sekhar, C Chandra, Murthy, Hema A

The two-pass information bottleneck (TPIB) based speaker diarization system operates independently on different conversational recordings. TPIB system does not consider previously learned speaker discriminative information while diarizing new convers

Externí odkaz: http://arxiv.org/abs/1902.08051

Zobrazit plný text záznamu