Výsledky vyhledávání

DiPCo — Dinner Party Corpus

Autor: Maarten Van Segbroeck, Roland Maas, Ksenia Kutsenko, Bjorn Hoffmeister, Cirenia Huerta, Jan Trmal, Xuewen Luo, Ahmed Zaid, Tinh Nguyen, Maurizio Omologo

Publikováno v: INTERSPEECH

We present a speech data corpus that simulates a "dinner party" scenario taking place in an everyday home environment. The corpus was created by recording multiple groups of four Amazon employee volunteers having a natural conversation in English aro

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d39b7dbcc8f7c368dab9fe7a0c14d0fa
https://doi.org/10.21437/interspeech.2020-2800

Zobrazit plný text záznamu

CHiME-6 Challenge: Tackling Multispeaker Speech Recognition for Unsegmented Recordings

Publikováno v: CHiME 2020-6th International Workshop on Speech Processing in Everyday Environments
CHiME 2020-6th International Workshop on Speech Processing in Everyday Environments, May 2020, Barcelona / Virtual, Spain

International audience; Following the success of the 1st, 2nd, 3rd, 4th and 5th CHiME challenges we organize the 6th CHiME Speech Separation and Recognition Challenge (CHiME-6). The new challenge revisits the previous CHiME-5 challenge and further co

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f23f9c49b95df37c06c48fa6001688a2
https://doi.org/10.21437/chime.2020-1

Zobrazit plný text záznamu

Multi-task self-supervised learning for Robust Speech Recognition

Autor: Santiago Pascual, Mirco Ravanelli, Jianyuan Zhong, Jan Trmal, Joao Monteiro, Pawel Swietojanski, Yoshua Bengio

Publikováno v: ICASSP

Despite the growing interest in unsupervised learning, extracting meaningful knowledge from unlabelled audio remains an open challenge. To take a step in this direction, we recently proposed a problem-agnostic speech encoder (PASE), that combines a c

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::99a19f66f756458a9da225e87dab1244
http://arxiv.org/abs/2001.09239

Zobrazit plný text záznamu

Zero-Shot Pronunciation Lexicons for Cross-Language Acoustic Model Transfer

Autor: David Yarowsky, Jan Trmal, Matthew Wiesner, Sanjeev Khudanpur, Oliver Adams

Publikováno v: ASRU

Existing acoustic models can be transferred to any language with a pronunciation lexicon (lexicon) that uses the same set of sub-word units as in training. Unfortunately such lexicons are not readily available in many low-resource languages. We bypas

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::5fdaac94e49f5d9195b888480b7242a9
https://doi.org/10.1109/asru46091.2019.9004019

Zobrazit plný text záznamu

Bottom-Up Unsupervised Word Discovery via Acoustic Units

Autor: Sanjeev Khudanpur, Jan Trmal, Chunxi Liu, Najim Dehak, Saurabhchand Bhati, Jesús Villalba

Publikováno v: GlobalSIP

Unsupervised term discovery is the task of identifying and grouping reoccurring word-like patterns from the untranscribed audio data. It facilitates unsupervised acoustic model training in zero resource setting where no or minimal transcribed speech

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::8fd2a7ca7ae4642b08321c71e8f09107
https://doi.org/10.1109/globalsip45357.2019.8969225

Zobrazit plný text záznamu

Using ASR Methods for OCR

Autor: Desh Raj, Chun Chieh Chang, Jan Trmal, Hossein Hadian, Yiwen Shao, Sanjeev Khudanpur, Babak Rekabdar, Paola Garcia, Bagher BabaAli, David Etter, Shinji Watanabe, Vimal Manohar, Daniel Povey, Ashish Arora

Publikováno v: ICDAR

Hybrid deep neural network hidden Markov models (DNN-HMM) have achieved impressive results on large vocabulary continuous speech recognition (LVCSR) tasks. However, the recent approaches using DNN-HMM models are not explored much for text recognition

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::b3d6e0662c2fbd501b6f86ac15695741
https://doi.org/10.1109/icdar.2019.00111

Zobrazit plný text záznamu

Induced Inflection-Set Keyword Search in Speech

Autor: Matthew Wiesner, Jan Trmal, Garrett Nicolai, David Yarowsky, Oliver Adams

Publikováno v: SIGMORPHON

We investigate the problem of searching for a lexeme-set in speech by searching for its inflectional variants. Experimental results indicate how lexeme-set search performance changes with the number of hypothesized inflections, while ablation experim

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c3482004361c5eceefb64568acb218de

Zobrazit plný text záznamu

Improving LF-MMI Using Unconstrained Supervisions for ASR

Autor: Jan Trmal, Hossein Sameti, Hossein Hadian, Daniel Povey, Sanjeev Khudanpur

Publikováno v: SLT

We present our work on improving the numerator graph for discriminative training using the lattice-free maximum mutual information (MMI) criterion. Specifically, we propose a scheme for creating unconstrained numerator graphs by removing time constra

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::e2ae2ebf5d41399c69b824fcc70d2ca3
https://doi.org/10.1109/slt.2018.8639684

Zobrazit plný text záznamu

The fifth 'CHiME' Speech Separation and Recognition Challenge: Dataset, task and baselines

Autor: Jon Barker, Jan Trmal, Shinji Watanabe, Emmanuel Vincent

Publikováno v: Interspeech 2018-19th Annual Conference of the International Speech Communication Association
Interspeech 2018-19th Annual Conference of the International Speech Communication Association, Sep 2018, Hyderabad, India
INTERSPEECH

International audience; The CHiME challenge series aims to advance robust automatic speech recognition (ASR) technology by promoting research at the interface of speech and language processing, signal processing , and machine learning. This paper int

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::40f3c5f936ca2eea2faffe993aa09f4a
https://hal.inria.fr/hal-01744021/document

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání