EESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based Decoding

Miao, Yajie; Gowayyed, Mohammad; Metze, Florian

doi:10.48550/arxiv.1507.08240

Cited by 36 publications

(19 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The landscape of neural network acoustic modeling is rapidly evolving. Spurred by the success of deep feed-forward neural nets for LVCSR in [1] and inspired by other research areas like image classification and natural language processing, many speech groups have looked at more sophisticated architectures such as deep convolutional nets [2,3], deep recurrent nets [4], time-delay neural nets [5], and long-short term memory nets [6,7,8,9]. The trend is to remove a lot of the complexity and human knowledge that was necessary in the past to build good ASR systems (e.g.…”

Section: Introductionmentioning

confidence: 99%

The IBM 2016 English Conversational Telephone Speech Recognition System

Saon¹,

Sercu²,

Rennie³

et al. 2016

Interspeech 2016

105

104

View full text Add to dashboard Cite

We describe a collection of acoustic and language modeling techniques that lowered the word error rate of our English conversational telephone LVCSR system to a record 6.6% on the Switchboard subset of the Hub5 2000 evaluation testset. On the acoustic side, we use a score fusion of three strong models: recurrent nets with maxout activations, very deep convolutional nets with 3x3 kernels, and bidirectional long short-term memory nets which operate on FMLLR and i-vector features. On the language modeling side, we use an updated model "M" and hierarchical neural network LMs.

show abstract

Section: Introductionmentioning

confidence: 99%

The IBM 2016 English Conversational Telephone Speech Recognition System

Saon¹,

Sercu²,

Rennie³

et al. 2016

Interspeech 2016

105

104

View full text Add to dashboard Cite

show abstract

“…In this paper, the contribution of the innovative research are summarised as below: [13][14][15][16][17][18][19][20]…”

Section: Discussionmentioning

confidence: 99%

“…Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks and recurrent neural networks have been developed. Deep learning has been successfully applied in the areas of speech recognition [15,16,24], image processing [17], object detection [18,19], drug discovery [20] and genomics [21], etc. Especially, deep convolutional nets (ConvNets) have brought about breakthroughs in processing images [22], video [23], speech [24] and audio [25], whereas recurrent nets have shone light on sequential data such as text and speech [26].…”

Section: Existing Work In Deep Learningmentioning

confidence: 99%

A Heuristically Self-Organised Linguistic Attribute Deep Learning in Edge Computing For IoT Intelligence

He,

Zhu

2020

Preprint

View full text Add to dashboard Cite

With the development of Internet of Things (IoT), IoT intelligence becomes emerging technology. "Curse of Dimensionality" is the barrier of data fusion in edge devices for the success of IoT intelligence. Deep learning has attracted great attention recently, due to the successful applications in several areas, such as image processing and natural language processing. However, the success of deep learning benefits from GPU computing. A Linguistic Attribute Hierarchy (LAH), embedded with Linguistic Decision Trees (LDTs) can represent a new attribute deep learning. In contrast to the conventional deep learning, an LAH could overcome the shortcoming of missing interpretation by providing transparent information propagation through the rules, produced by LDTs in the LAH. Similar to the conventional deep learning, the computing complexity of optimising LAHs blocks the applications of LAHs.In this paper, we propose a heuristic approach to constructing an LAH, embedded with LDTs for decision making or classification by utilising the distance correlations between attributes and between attributes and the goal variable. The set of attributes is divided to some attribute clusters, and then they are heuristically organised to form a linguistic attribute hierarchy. The proposed approach was validated with some benchmark decision making or classification problems from the UCI machine learning repository. The experimental results show that the proposed self-organisation algorithm can construct an effective and efficient linguistic attribute hierarchy. Such a self-organised linguistic attribute hierarchy embedded with LDTs can not only efficiently tackle 'curse of dimensionality' in a single LDT for data fusion with massive attributes, but also achieve better or comparable performance on decision making or classification, compared to the single LDT for the problem to be solved. The self-organisation algorithm is much efficient than the Genetic Algorithm in Wrapper for the optimisation of LAHs. This makes it feasible to embed the self-organisation algorithm in edge devices for IoT intelligence.

show abstract

“…We next evaluate our models on Wall Street Journal (WSJ) corpus (available as LDC corpus LDC93S6B and LDC94S13B), where we use the full 81 hour set "si284" for training, set "dev93" for validation and set "eval92" for test. We follow the same data preparation process and model setting as in [18], and we use 59 characters as the targets for the acoustic modelling. Decoding is done with the CTC [21] based weighted finite-state transducers (WFSTs) [22] as proposed by [18].…”

Section: Speech Recognitionmentioning

confidence: 99%

“…We follow the same data preparation process and model setting as in [18], and we use 59 characters as the targets for the acoustic modelling. Decoding is done with the CTC [21] based weighted finite-state transducers (WFSTs) [22] as proposed by [18].…”

Section: Speech Recognitionmentioning

confidence: 99%

On Multiplicative Integration with Recurrent Neural Networks

Wu¹,

Zhang²,

Zhang³

et al. 2016

Preprint

View full text Add to dashboard Cite

We introduce a general and simple structural design called "Multiplicative Integration" (MI) to improve recurrent neural networks (RNNs). MI changes the way in which information from difference sources flows and is integrated in the computational building block of an RNN, while introducing almost no extra parameters. The new structure can be easily embedded into many popular RNN models, including LSTMs and GRUs. We empirically analyze its learning behaviour and conduct evaluations on several tasks using different RNN models. Our experimental results demonstrate that Multiplicative Integration can provide a substantial performance boost over many of the existing RNN models.

show abstract

EESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based Decoding

Cited by 36 publications

References 25 publications

The IBM 2016 English Conversational Telephone Speech Recognition System

The IBM 2016 English Conversational Telephone Speech Recognition System

A Heuristically Self-Organised Linguistic Attribute Deep Learning in Edge Computing For IoT Intelligence

On Multiplicative Integration with Recurrent Neural Networks

Contact Info

Product

Resources

About