Towards Better Decoding and Language Model Integration in Sequence to Sequence Models
Open accessThe DOI matches its Crossref record; 47 of the 47 references it cites were found in a registry or an index; none of the 9 sources its claims cite is checked for support yet.
Abstract
The recently proposed Sequence-to-Sequence (seq2seq) framework advocates replacing complex data processing pipelines, such as an entire automatic speech recognition system, with a single neural network trained in an end-to-end fashion.In this contribution, we analyse an attention-based seq2seq speech recognition system that directly transcribes recordings into characters.We observe two shortcomings: overconfidence in its predictions and a tendency to produce incomplete transcriptions when language models are used.We propose practical solutions to both problems achieving competitive speaker independent word error rates on the Wall Street Journal dataset: without separate language models we reach 10.6% WER, while together with a trigram language model, we reach 6.7% WER.
Overview
CoChat Summary
TL;DR
This paper analyzes two failure modes of an attention-based seq2seq speech recognizer: overconfident character predictions and incomplete transcripts when using language models. It proposes label smoothing plus decode-time coverage handling, reaching 10.6% WER without a separate language model and 6.7% WER with a trigram language model on WSJ.
Key findings
- The authors report that the baseline seq2seq model on WSJ achieved 14.2% WER on eval92, while unigram label smoothing reduced this to 10.6% and temporal/neighborhood smoothing to 10.7% on eval92.
- With an extended trigram language model and coverage, the best reported result was 6.7% WER on eval92; the corresponding dev93 score was 9.7% for temporal label smoothing + coverage.
- They found that model overconfidence makes beam search less effective and that increasing SoftMax temperature can help beam search, but without changing greedy decoding accuracy.
- For incomplete transcripts, simply constraining EOS was not sufficient at large beam widths; promoting coverage helped, while maximizing transcript length alone caused more word insertions and worse overall WER.
Methods
- Attention-based seq2seq speech recognition model built on Listen, Attend and Spell: a multilayer Bi-LSTM listener, an LSTM speller, and attention.
- Experiments were run on the Wall Street Journal dataset, training on si284, validating on dev93, and evaluating on eval92; inputs were 80-dimensional mel-scale filterbanks with first and second differences and per-speaker normalization.
- Main analysis compared decoding strategies and regularizers: SoftMax temperature changes, label smoothing variants (uniform/unigram and neighborhood/temporal), beam search settings, and a decode-time coverage penalty with an extended trigram language model.
Limitations
- The paper states that adding a language model introduces two challenges: model overconfidence makes balancing the beam-search terms difficult, and incomplete transcripts occur unless a coverage term is added.
- The authors note that simply maximizing transcript length yields more word insertion errors and worse overall WER.
- Evident from the text: results are reported on a single dataset (WSJ) with a particular ASR setup, so generalization beyond this setting is not established in the paper.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 2 Not valid · 7 Not checked · 37 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We observe two shortcomings: overconfidence in its predictions and a tendency to produce incomplete transcriptions when language models are used.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A width 100 beam search with a trigram language model finds the second transcript, which misses the beginning of the utterance.”
plainstates a causeSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we didn't observe consistent gains from this extension of the beam search criterion.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“| Temporal LS | 6.6M | 14.1 | 10.7 |”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Second, incomplete transcripts were produced unless a recording coverage term was added.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
20 more claims
Create a free account to see all 46 claims and to verify the 9 references still open.
- It may state two separate claims as one. 16
- Probably not a substantial claim. 5
- The quote only partly states this claim. 2
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“Deep learning [1] has led to many breakthroughs including speech and image recognition [2, 3, 4, 5, 6, 7].”
BackgroundAdam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“Networks were trained using 8 asynchronous replica workers each employing the ADAM algorithm [21] with default parameters and the learning rate set initially to 10−3, then reduced to 10−4 and 10−5 after 400k and 500k training steps, respectively.”
MethodRethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766 · cited 1×
“Especially, Sequence-to-Sequence (seq2seq) neural networks have proved to be very successful on complex transduction tasks, such as machine translation [8, 9, 10], speech recognition [11, 12, 13], or lip-reading [14].”
BackgroundNeural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213 · cited 1×
“Especially, Sequence-to-Sequence (seq2seq) neural networks have proved to be very successful on complex transduction tasks, such as machine translation [8, 9, 10], speech recognition [11, 12, 13], or lip-reading [14].”
BackgroundHinton, G., Deng, L., Yu, D., Dahl, G., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V., Nguyen, P., Sainath, T., et al. (2012) · IEEE Signal Processing Magazine · doi:10.1109/msp.2012.2205597 · Cited by 10,260
Self-citationTensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408
Effective Approaches to Attention-based Neural Machine Translation
Luong, T., Pham, H., & Manning, C. D. (2015) · Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d15-1166 · Cited by 8,382 · cited 1×
“Past attentions vectors were used as auxiliary inputs in the emitting RNN either directly [34], or as cumulative coverage information [35].”
MethodAsynchronous Methods for Deep Reinforcement Learning
Mnih, V., Badia, A. P., Mirza, M., Graves, A., Lillicrap, T. P., Harley, T., Silver, D., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1602.01783 · Cited by 7,132 · cited 1×
“Smoothing techniques increase the entropy of a model’s predictions, a technique that was used to promote exploration in reinforcement learning [27, 28, 29].”
MethodGoogle's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319 · cited 2×
“Our solution that employs a coverage penalty at decode time only is most similar to the one used by the Google Translation system [10].”
BackgroundMethodKaldi Speech Recognition Toolkit
Povey, D. (2024) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · doi:10.57702/jb3fvbn9 · Cited by 4,609
Towards End-To-End Speech Recognition with Recurrent Neural Networks
Graves, A., & Jaitly, N. (2014) · Cited by 2,557
Self-citationNeural Networks: Tricks of the Trade: Second Edition
(2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8 · Cited by 2,116
Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
Amodei, D., Anubhai, R., Battenberg, E., Case, C. T., Casper, J., Catanzaro, B., Chen, J., Chrzanowski, M., Coates, A., Diamos, G., et al. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1512.02595 · Cited by 2,110 · cited 1×
“Deep learning [1] has led to many breakthroughs including speech and image recognition [2, 3, 4, 5, 6, 7].”
BackgroundAttention-Based Models for Speech Recognition
Chorowski, J., Bahdanau, D., Serdyuk, D., Cho, K., & Bengio, Y. (2015) · arXiv · doi:10.48550/arxiv.1506.07503 · Cited by 2,021 · cited 2×
“The attention mechanism is sensitive to the location of frames selected during the previous step and employs the convolutional filters over the previous attention weights [11].”
BackgroundMethodSelf-citationLeCun, Y. A., Bottou, L., Orr, G. B., & Müller, K.-R. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8_3 · Cited by 1,264
End-to-end attention-based large vocabulary speech recognition
Bahdanau, D., Chorowski, J., Serdyuk, D., Brakel, P., & Bengio, Y. (2016) · 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2016.7472618 · Cited by 1,096 · cited 6×
“Language Model Integration during decoding can be implemented by extending the beam search cost [12, 4, 15]:”
BackgroundMethodInfluentialSelf-citationRethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2015) · arXiv · doi:10.48550/arxiv.1512.00567 · Cited by 741 · cited 4×
“Label smoothing prevents saturating the SoftMax nonlinearity and results in better gradient flow to lower layers of the network [16].”
BackgroundMethodInfluentialEESEN: End-to-end speech recognition using deep RNN models and WFST-based decoding
Miao, Y., Gowayyed, M., & Metze, F. (2015) · 2015 IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru.2015.7404790 · Cited by 707 · cited 3×
“Our best result surpasses CTCbased networks [6] and matches the results of a DNN-HMM and CTC ensemble [23].”
BackgroundMethodResultLip reading sentences in the wild
Chung, J. S., Senior, A., Vinyals, O., & Zisserman, A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.367 · Cited by 702 · cited 1×
“Especially, Sequence-to-Sequence (seq2seq) neural networks have proved to be very successful on complex transduction tasks, such as machine translation [8, 9, 10], speech recognition [11, 12, 13], or lip-reading [14].”
BackgroundModeling Coverage for Neural Machine Translation
Tu, Z., Lu, Z., Liu, Y., Liu, X., & Li, H. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1008 · Cited by 692 · cited 1×
“Past attentions vectors were used as auxiliary inputs in the emitting RNN either directly [34], or as cumulative coverage information [35].”
MethodEffective Approaches to Attention-based Neural Machine Translation
Luong, M.-T., Pham, H., & Manning, C. D. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1508.04025 · Cited by 655 · cited 1×
“Past attentions vectors were used as auxiliary inputs in the emitting RNN either directly [34], or as cumulative coverage information [35].”
MethodRegularizing Neural Networks by Penalizing Confident Output Distributions
Pereyra, G., Tucker, G., Chorowski, J., Kaiser, Ł., & Hinton, G. (2017) · arXiv · doi:10.48550/arxiv.1701.06548 · Cited by 655
Self-citationOpenFst: A General and Efficient Weighted Finite-State Transducer Library
Allauzen, C., Riley, M., Schalkwyk, J., Skut, W., & Mohri, M. (2007) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-76336-9_3 · Cited by 539 · cited 1×
“We have use the FST framework to compose the language model with a "spelling lexicon" [6, 12, 19].”
MethodSequence-to-Sequence Learning as Beam-Search Optimization
Wiseman, S., & Rush, A. M. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1137 · Cited by 481 · cited 1×
“Global normalization has yielded important benefits on many NLP tasks including parsing and translation [32, 33].”
BackgroundGlobally Normalized Transition-Based Neural Networks
Andor, D., Alberti, C., Weiss, D., Severyn, A., Presta, A., Ganchev, K., Petrov, S., & Collins, M. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1231 · Cited by 472 · cited 1×
“Global normalization has yielded important benefits on many NLP tasks including parsing and translation [32, 33].”
BackgroundOn Using Monolingual Corpora in Neural Machine Translation
Gülçehre, Ç., Fırat, O., Xu, K., Cho, K., Barrault, L., Lin, H.-C., Bougares, F., Schwenk, H., & Bengio, Y. (2015) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.48550/arxiv.1503.03535 · Cited by 466 · cited 1×
“Language Model Integration during decoding can be implemented by extending the beam search cost [12, 4, 15]:”
BackgroundVery deep convolutional networks for end-to-end speech recognition
Zhang, Y., Chan, W., & Jaitly, N. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7953077 · Cited by 407 · cited 1×
“Label smoothing brings a large error rate reduction, nearly matching the performance achieved with very deep and sophisticated encoders [22].”
BackgroundSelf-citationFunction Optimization using Connectionist Reinforcement Learning Algorithms
WILLIAMS, R. J., & PENG, J. (1991) · Connection Science · doi:10.1080/09540099108946587 · Cited by 288 · cited 1×
“Smoothing techniques increase the entropy of a model’s predictions, a technique that was used to promote exploration in reinforcement learning [27, 28, 29].”
MethodChan, W., Jaitly, N., Le, Q. V., & Vinyals, O. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1508.01211 · Cited by 284
Self-citationDisturbLabel: Regularizing CNN on the Loss Layer
Xie, L., Wang, J., Wei, Z., Wang, M., & Tian, Q. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.514 · Cited by 200 · cited 1×
“Several improved smoothing schemes were proposed, including sampling erroneous labels instead of using a fixed distribution [25], using the marginal label probabilities [17], or using early errors of the model [26].”
MethodEESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based DecodingPreprint
Miao, Y., Gowayyed, M., & Metze, F. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1507.08240 · Cited by 164 · cited 3×
“Our best result surpasses CTCbased networks [6] and matches the results of a DNN-HMM and CTC ensemble [23].”
BackgroundMethodResultModeling Coverage for Neural Machine TranslationPreprint
Tu, Z., Lu, Z., Liu, Y., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1601.04811 · Cited by 157 · cited 1×
“Past attentions vectors were used as auxiliary inputs in the emitting RNN either directly [34], or as cumulative coverage information [35].”
MethodDiscriminative learning in sequential pattern recognition
He, X., Deng, L., & Chou, W. (2008) · IEEE Signal Processing Magazine · doi:10.1109/msp.2008.926652 · Cited by 143 · cited 1×
“In discriminative training of classical ASR systems normalization is performed over lattices [31].”
Sequence-to-Sequence Learning as Beam-Search OptimizationPreprint
Wiseman, S., & Rush, A. M. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.02960 · Cited by 113 · cited 1×
“Global normalization has yielded important benefits on many NLP tasks including parsing and translation [32, 33].”
BackgroundLatent Sequence DecompositionsPreprint
Chan, W., Zhang, Y., Le, Q. V., & Jaitly, N. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.03035 · Cited by 61
Self-citationLearning online alignments with continuous rewards policy gradient
Luo, Y., Chiu, C., Jaitly, N., & Sutskever, I. (2017) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2017.7952667 · Cited by 44 · cited 1×
“Smoothing techniques increase the entropy of a model’s predictions, a technique that was used to promote exploration in reinforcement learning [27, 28, 29].”
MethodSelf-citationCoverage Embedding Models for Neural Machine TranslationPreprint
Mi, H., Sankaran, B., Wang, Z., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.03148 · Cited by 35 · cited 1×
“Coverage embeddings vectors associated with source words end modified during training were proposed in [36].”
BackgroundA Coverage Embedding Model for Neural Machine Translation.Preprint
Mi, H., Sankaran, B., Wang, Z., & Ittycheriah, A. (2016) · arXiv (Cornell University) · Cited by 21
SoftTarget Regularization: An Effective Technique to Reduce Over-Fitting in Neural Networks
Aghajanyan, A. (2017) · IEEE International Conference on Cybernetics (CYBCONF) · doi:10.1109/cybconf.2017.7985811 · Cited by 18 · cited 1×
“Several improved smoothing schemes were proposed, including sampling erroneous labels instead of using a fixed distribution [25], using the marginal label probabilities [17], or using early errors of the model [26].”
MethodLearning Online Alignments with Continuous Rewards Policy GradientPreprint
Luo, Y., Chiu, C., Jaitly, N., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.01281 · Cited by 10 · cited 1×
“Smoothing techniques increase the entropy of a model’s predictions, a technique that was used to promote exploration in reinforcement learning [27, 28, 29].”
MethodSelf-citationVery Deep Convolutional Networks for End-to-End Speech RecognitionPreprint
Zhang, Y., Chan, W., & Jaitly, N. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.03022 · Cited by 10 · cited 1×
“Label smoothing brings a large error rate reduction, nearly matching the performance achieved with very deep and sophisticated encoders [22].”
BackgroundSelf-citationDisturbLabel: Regularizing CNN on the Loss LayerPreprint
Xie, L., Jingdong, W., Wei, Z., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.00055 · Cited by 2 · cited 1×
“Several improved smoothing schemes were proposed, including sampling erroneous labels instead of using a fixed distribution [25], using the marginal label probabilities [17], or using early errors of the model [26].”
Method(2014) · doi:10.4230/oasics.slate.2021.16 · Cited by 1 · cited 5×
“Language Model Integration during decoding can be implemented by extending the beam search cost [12, 4, 15]:”
BackgroundInfluentialGlobally Normalized Transition-Based Neural Networks
Andor, D., Alberti, C., Weiss, D., Severyn, A., Presta, A., Ganchev, K., Petrov, S., & Collins, M. (2016) · arXiv · doi:10.48550/arxiv.1603.06042 · cited 1×
“Global normalization has yielded important benefits on many NLP tasks including parsing and translation [32, 33].”
BackgroundSoftTarget Regularization: An Effective Technique to Reduce Over-Fitting in Neural Networks
Aghajanyan, A. (2016) · arXiv · doi:10.48550/arxiv.1609.06693 · cited 1×
“Several improved smoothing schemes were proposed, including sampling erroneous labels instead of using a fixed distribution [25], using the marginal label probabilities [17], or using early errors of the model [26].”
MethodLip Reading Sentences in the Wild
Chung, J. S., Senior, A., Vinyals, O., & Zisserman, A. (2016) · arXiv · doi:10.48550/arxiv.1611.05358 · cited 1×
“Especially, Sequence-to-Sequence (seq2seq) neural networks have proved to be very successful on complex transduction tasks, such as machine translation [8, 9, 10], speech recognition [11, 12, 13], or lip-reading [14].”
Background
Research works printed but not linked: 2
- [1][1] Y. LeCun, Y. Bengio, and G. Hinton, 'Deep learning,' Nature , vol. 521, no. 7553, pp. 436-444, 2015. -
- [23][23] A. Graves and N. Jaitly, 'Towards End-To-End Speech Recognition with Recurrent Neural Networks,' in ICML '14 , 2014, pp. 17641772. -
Cited by
SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
Park, D. S., Chan, W., Zhang, Y., Chiu, C.-C., Zoph, B., Cubuk, E. D., & Le, Q. V. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2680 · Cited by 3,558
“We also use a coverage penalty c [29].”
MethodHierarchical Neural Story Generation
Fan, A., Lewis, M., & Dauphin, Y. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1082 · Cited by 1,266
“…together without architectural modifications: Ramachandran et al. (2016) use language models to initialize the encoder and decoder side of the seq2seq model independently, and Chorowski and Jaitly (2016) combine the predictions of the language model and seq2seq model solely at inference time.”
MethodHierarchical Neural Story Generation
Fan, A., Lewis, M., & Dauphin, Y. (2018) · arXiv · doi:10.48550/arxiv.1805.04833 · Cited by 1,242
“…together without architectural modifications: Ramachandran et al. (2016) use language models to initialize the encoder and decoder side of the seq2seq model independently, and Chorowski and Jaitly (2016) combine the predictions of the language model and seq2seq model solely at inference time.”
MethodSpeech-Transformer: A No-Recurrence Sequence-to-Sequence Model for Speech Recognition
Dong, L., Xu, S., & Xu, B. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8462506 · Cited by 1,100
State-of-the-Art Speech Recognition with Sequence-to-Sequence Models
Chiu, C.-C., Sainath, T. N., Wu, Y., Prabhavalkar, R., Nguyen, P., Chen, Z., Kannan, A., Weiss, R. J., Rao, K., Gonina, E., et al. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8462105 · Cited by 1,043
Self-citation
Metadata
- Authors
- Chorowski, J., & Jaitly, N.
- Affiliations
- Google (United States) (US)
- ORCID
- Chorowski, Jan
- Venue
- Interspeech 2017
- Publisher
- ISCA
- Published
- 2017-08-20
- Pages
- 523-527
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W2577366047
- Cited by
- 358
- References
- 34
- Citations by year
- 2026: 4 · 2025: 14 · 2024: 13 · 2023: 39 · 2022: 26 · 2021: 73 · 2020: 88 · 2019: 49 · 2018: 40 · 2017: 12
- FWCI
- 23.58
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Interspeech (USB) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Handwritten Text Recognition Techniques
- Keywords
- attention-based models, language model integration, trigram language model, end-to-end speech recognition, speaker-independent recognition, Wall Street Journal
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.