检索结果-内蒙古大学图书馆

Topic identification of spoken documents using unsupervised acoustic unit discovery

学校读者我要写书评

暂无评论

Topic identification of spoken documents using unsupervised ...

2017 IEEE International Conference on Acoustics, speech, and Signal processing, ICASSP 2017

作者： Kesiraju, Santosh Pappagari, Raghavendra Ondel, Lucas Burget, Lukas Dehak, Najim Khudanpur, Sanjeev Černocký, Jan Gangashetty, Suryakanth V Brno University of Technology SpeechatFIT and IT4I Center of Excellence Czech Republic Center for Language and Speech Processing Johns Hopkins University Baltimore United States International Institute of Information Technology Hyderabad India

ISBN: (纸本)9781509041176

This paper investigates the application of unsupervised acoustic unit discovery for topic identification (topic ID) of spoken audio documents. The acoustic unit discovery method is based on a non-parametric Bayesian phone-loop model that segments a speech utterance into phone-like categories. The discovered phone-like (acoustic) units are further fed into the conventional topic ID framework. Using multilingual bottleneck features for the acoustic unit discovery, we show that the proposed method outperforms other systems that are based on cross-lingual phoneme recognizer. © 2017 IEEE.

关键词： acoustic unit discovery non-parametric Bayesian models topic identification unsupervised learning

Acoustic data-driven lexicon learning based on a greedy pronunciation selection framework

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zhang, Xiaohui Manohar, Vimal Povey, Daniel Khudanpur, Sanjeev Center for Language and Speech Processing Human Language Technology Center of Excellence Johns Hopkins University BaltimoreMD21218 United States

speech recognition systems for irregularly-spelled languages like English normally require hand-written pronunciations. In this paper, we describe a system for automatically obtaining pronunciations of words for which pronunciations are not available, but for which transcribed data exists. Our method integrates information from the letter sequence and from the acoustic evidence. The novel aspect of the problem that we address is the problem of how to prune entries from such a lexicon (since, empirically, lexicons with too many entries do not tend to be good for ASR performance). Experiments on various ASR tasks show that, with the proposed framework, starting with an initial lexicon of several thousand words, we are able to learn a lexicon which performs close to a full expert lexicon in terms of WER performance on test data, and is better than lexicons built using G2P alone or with a pruning criterion based on pronunciation probability. Copyright © 2017, The Authors. All rights reserved.

关键词： speech recognition

Using of heterogeneous corpora for training of an ASR system

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Trmal, Jan Kumar, Gaurav Manohar, Vimal Khudanpur, Sanjeev Post, Matt McNamee, Paul Center for Language and Speech Processing Human Language Technology Center of Excellence Johns Hopkins University BaltimoreMD21218 United States

The paper summarizes the development of the LVCSR system built as a part of the Pashto speech-translation system at the SCALE (Summer Camp for Applied language Exploration) 2015 workshop on "speech-to-text-translation for low-resource languages". The Pashto language was chosen as a good "proxy" low-resource language, exhibiting multiple phenomena which make the speech-recognition and and speech-to-text-translation systems development hard. Even when the amount of data is seemingly sufficient, given the fact that the data originates from multiple sources, the preliminary experiments reveal that there is little to no benefit in merging (concatenating) the corpora and more elaborate ways of making use of all of the data must be worked out. This paper concentrates only on the LVCSR part and presents a range of different techniques that were found to be useful in order to benefit from multiple different corpora. Copyright © 2017, The Authors. All rights reserved.

关键词： speech recognition

Grammatical error correction with neural reinforcement learning

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Sakaguchi, Keisuke Post, Matt van Durme, Benjamin Center for Language and Speech Processing Johns Hopkins University United States Human Language Technology Center of Excellence Johns Hopkins University United States

We propose a neural encoder-decoder model with reinforcement learning (NRL) for grammatical error correction (GEC). Unlike conventional maximum likelihood estimation (MLE), the model directly optimizes towards an objective that considers a sentence-level, task-specific evaluation metric, avoiding the exposure bias issue in MLE. We demonstrate that NRL outperforms MLE both in human and automated evaluation metrics, achieving the state-ofthe-art on a fluency-oriented GEC corpus. Copyright © 2017, The Authors. All rights reserved.

关键词： Error correction

Topic identification for speech without ASR

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Liu, Chunxi Trmal, Jan Wiesner, Matthew Harman, Craig Khudanpur, Sanjeev Center for Language and Speech Processing Johns Hopkins University United States Human Language Technology Center of Excellence Johns Hopkins University United States

Modern topic identification (topic ID) systems for speech use automatic speech recognition (ASR) to produce speech transcripts, and perform supervised classification on such ASR outputs. However, under resource-limited conditions, the manually transcribed speech required to develop standard ASR systems can be severely limited or unavailable. In this paper, we investigate alternative unsupervised solutions to obtaining tokenizations of speech in terms of a vocabulary of automatically discovered word-like or phoneme-like units, without depending on the supervised training of ASR systems. Moreover, using automatic phoneme-like tokenizations, we demonstrate that a convolutional neural network based framework for learning spoken document representations provides competitive performance compared to a standard bag-of-words representation, as evidenced by comprehensive topic ID evaluations on both single-label and multi-label classification tasks. Copyright © 2017, The Authors. All rights reserved.

关键词： speech

Feature Generation for Robust Semantic Role Labeling

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Wolfe, Travis Dredze, Mark van Durme, Benjamin Human Language Technology Center of Excellence Johns Hopkins University

Hand-engineered feature sets are a well understood method for creating robust NLP models, but they require a lot of expertise and effort to create. In this work we describe how to automatically generate rich feature sets from simple units called featlets, requiring less engineering. Using information gain to guide the generation process, we train models which rival the state of the art on two standard Semantic Role Labeling datasets with almost no task or linguistic insight. Copyright © 2017, The Authors. All rights reserved.

关键词： Semantics

The JHU Machine Translation Systems for WMT 2016 1

学校读者我要写书评

暂无评论

The JHU Machine Translation Systems for WMT 2016

1st Conference on Machine Translation, WMT 2016, held at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016

作者： Ding, Shuoyang Duh, Kevin Khayrallah, Huda Koehn, Philipp Post, Matt Center for Language and Speech Processing Human Language Technology Center of Excellence Department of Computer Science Johns Hopkins University BaltimoreMD United States

ISBN: (纸本)9781945626104

This paper describes the submission of Johns Hopkins University for the shared translation task of ACL 2016 First Conference on Machine Translation (WMT 2016). We set up phrase-based, hierarchical phrase-based and syntax-based systems for all 12 language pairs of this year's evaluation campaign. Novel research directions we investigated include: neural probabilistic language models, bilingual neural network language models, morphological segmentation, and the attentionbased neural machine translation model as reranking feature. © 2016 Association for Computational Linguistics.

关键词： Neural machine translation

Joint acoustic and class inference for weakly supervised sound event detection

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Kothinti, Sandeep Imoto, Keisuke Chakrabarty, Debmalya Sell, Gregory Watanabe, Shinji Elhilali, Mounya Department of Electrical and Computer Engineering Johns Hopkins University BaltimoreMD United States College of Information Science and Engineering Ritsumeikan University Shiga Japan Human Language Technology Center of Excellence Johns Hopkins University BaltimoreMD United States

Sound event detection is a challenging task, especially for scenes with multiple simultaneous events. While event classification methods tend to be fairly accurate, event localization presents additional challenges, especially when large amounts of labeled data are not available. Task4 of the 2018 DCASE challenge presents an event detection task that requires accuracy in both segmentation and recognition of events while providing only weakly labeled training data. Supervised methods can produce accurate event labels but are limited in event segmentation when training data lacks event timestamps. On the other hand, unsupervised methods that model the acoustic properties of the audio can produce accurate event boundaries but are not guided by the characteristics of event classes and sound categories. We present a hybrid approach that combines an acoustic-driven event boundary detection and a supervised label inference using a deep neural network. This framework leverages benefits of both unsupervised and supervised methodologies and takes advantage of large amounts of unlabeled data, making it ideal for large-scale weakly labeled event detection. Compared to a baseline system, the proposed approach delivers a 15% absolute improvement in F-score, demonstrating the benefits of the hybrid bottom-up, top-down approach. Copyright © 2018, The Authors. All rights reserved.

关键词： Acoustic properties

Large language models surpass human experts in predicting neuroscience results

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Luo, Xiaoliang Rechardt, Akilles Sun, Guangzhi Nejad, Kevin K. Yáñez, Felipe Yilmaz, Bati Lee, Kangjoo Cohen, Alexandra O. Borghesani, Valentina Pashkov, Anton Marinazzo, Daniele Nicholas, Jonathan Salatiello, Alessandro Sucholutsky, Ilia Minervini, Pasquale Razavi, Sepehr Rocca, Roberta Yusifov, Elkhan Okalova, Tereza Gu, Nianlong Ferianc, Martin Khona, Mikail Patil, Kaustubh R. Lee, Pui-Shee Mata, Rui Myers, Nicholas E. Bizley, Jennifer K. Musslick, Sebastian Bilgin, Isil Poyraz Niso, Guiomar Ales, Justin M. Gaebler, Michael Murty, N. Apurva Ratan Loued-Khenissi, Leyla Behler, Anna Hall, Chloe M. Dafflon, Jessica Bao, Sherry Dongqi Love, Bradley C. Department of Experimental Psychology University College London London United Kingdom Department of Engineering University of Cambridge Cambridge United Kingdom Department of Physiology Anatomy & Genetics University of Oxford Oxford United Kingdom Max Planck Institute for Neurobiology of Behavior – caesar Bonn Germany Bilkent University Ankara Turkey Department of Computer Science University of Bristol Bristol United Kingdom The Alan Turing Institute London United Kingdom Department of Psychiatry Yale University New Haven United States Psychology Emory University Atlanta United States Faculty of Psychology and Educational Sciences Université de Genève Genève Switzerland Neurosurgery Novosibirsk State Medical University Novosibirsk Russia Federal Center of Neurosurgery FSBI Novosibirsk Russia Department of Data Collection and Processing Systems Novosibirsk State Technical University Novosibirsk Russia Department of Data Analysis Ghent University Ghent Belgium Psychology New York University New York United States Department of Cognitive Neurology University of Tübingen Tübingen Germany Computer Science Princeton University Princeton United States ILCC University of Edinburgh Edinburgh United Kingdom Philosophy Psychology and Language Sciences The University of Edinburgh Edinburgh United Kingdom Department of Culture Cognition and Computation Aarhus University Aarhus Denmark Department of Molecular Life Sciences University of Zurich Zurich Switzerland Bioengineering University of Pennsylvania Philadelphia United States Linguistic Research Infrastructure University of Zurich Zurich Switzerland Electronic and Electrical Engineering University College London London United Kingdom Brain and Cognitive Sciences Massachusetts Institute of Technology CambridgeMA United States Institute of Neuroscience and Medicine INM-7: Brain and Behaviour Research Centre Jülich Jülich Germany Medical Faculty Institute of Systems Neuroscience Heinrich Hein

Scientific discoveries often hinge on synthesizing decades of research, a task that potentially outstrips human information processing capacities. Large language models (LLMs) offer a solution. LLMs trained on the vast scientific literature could potentially integrate noisy yet interrelated findings to forecast novel results better than human experts. To evaluate this possibility, we created BrainBench, a forward-looking benchmark for predicting neuroscience results. We find that LLMs surpass experts in predicting experimental outcomes. BrainGPT, an LLM we tuned on the neuroscience literature, performed better yet. Like human experts, when LLMs were confident in their predictions, they were more likely to be correct, which presages a future where humans and LLMs team together to make discoveries. Our approach is not neuroscience-specific and is transferable to other knowledge-intensive endeavors. © 2024, CC BY.

关键词： Prediction models