Japanese and Korean voice search
The DOI matches its Crossref record; 7 of the 8 references it cites were found in a registry or an index.
Abstract
This paper describes challenges and solutions for building a successful voice search system as applied to Japanese and Korean at Google. We describe the techniques used to deal with an infinite vocabulary, how modeling completely in the written domain for language model and dictionary can avoid some system complexity, and how we built dictionaries, language and acoustic models in this framework. We show how to deal with the difficulty of scoring results for multiple script languages because of ambiguities. The development of voice search for these languages led to a significant simplification of the original process to build a system for any new language which in in parts became our default process for internationalization of voice search.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“LogRobust [8], which encodes log templates into semantic vectors using the FastText pre-trained model [31], cannot work well on 2 out of 4 datasets.”
From Log-based Anomaly Detection Without Log Parsing (2021) · cites this paper as [8]
In the citing paper
“Their empirical study on a Microsoft online service system shows that up to 30.3% logs are changed in the latest version.”
From Log-based Anomaly Detection Without Log Parsing (2021) · cites this paper as [8]
In the citing paper
“Zhang et al. [8] indicated that log data is unstable, meaning that new log events often appear due to software evolution at its lifetime.”
From Log-based Anomaly Detection Without Log Parsing (2021) · cites this paper as [8]
In the citing paper
“Another option is to use domain knowledge to reduce the input space of a model — such strategies include patch embeddings [Dosovitskiy et al. 2020] as well as fixed or learned tokenization [Schuster and Nakajima 2012; Sennrich et al. 2016; Kudo and Richardson 2018; van den Oord et al. 2017; Ramesh et al. 2021]. These methods may alleviate some challenges facing generative approaches, but have the”
From On the Opportunities and Risks of Foundation Models (2021) · cites this paper as Schuster and Nakajima 2012
Citations
Cites
Semi-tied covariance matrices for hidden Markov models
Gales, M. J. F. (1999) · IEEE Transactions on Speech and Audio Processing · doi:10.1109/89.759034 · Cited by 537
Boosted MMI for model and feature-space discriminative training
Povey, D., Kanevsky, D., Kingsbury, B., Ramabhadran, B., Saon, G., & Visweswariah, K. (2008) · 2008 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2008.4518545 · Cited by 341
“Your Word is my Command”: Google Search by Voice: A Case Study
Schalkwyk, J., Beeferman, D., Beaufays, F., Byrne, B., Chelba, C., Cohen, M., Kamvar, M., & Strope, B. (2010) · Advances in Speech Recognition · doi:10.1007/978-1-4419-5951-5_4 · Cited by 232
Building transcribed speech corpora quickly and cheaply for many languages
Hughes, T., Nakajima, K., Ha, L., Vasu, A., Moreno, P. J., & LeBeau, M. (2010) · Interspeech (USB) · doi:10.21437/interspeech.2010-551 · Cited by 75
Self-citationGoogle Search by Voice: A Case Study
Schalkwyk, J., Beeferman, D., Beaufays, F., Byrne, B., Chelba, C. I., Cohen, M., Garrett, M., & Strope, B. P. (2010) · Cited by 63
Query language modeling for voice search
Chelba, C. I., Schalkwyk, J., Brants, T., Ha, V., Harb, B., Neveitt, W., Parada, C., & Xu, P. (2010) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt.2010.5700834 · Cited by 28
Search by voice in Mandarin Chinese
Shan, J., Wu, G., Hu, Z., Tang, X., Jansche, M., & Moreno, P. J. (2010) · Interspeech (USB) · doi:10.21437/interspeech.2010-129 · Cited by 18
Research works printed but not linked: 1
- sung, 0, Deploying Search by Voice in Cantonese, Proc Interspeech 2011
Cited by
Items 701–800 of 933
Knowledge-Based Neural Pre-training for Intelligent Document Management
Margiotta, D., Croce, D., Rotoloni, M., Cacciamani, B., & Basili, R. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-08421-8_39 · Cited by 1
High Tech team at SemEval-2022 Task 6: Intended Sarcasm Detection for Arabic texts
Alami, H., Benlahbib, A., & Alami, A. (2022) · International Workshop on Semantic Evaluation (SemEval) · doi:10.18653/v1/2022.semeval-1.116 · Cited by 1
“We apply the pre-trained BERT tokenizer which is based on wordpiece model (Schuster and Nakajima, 2012).”
MethodAuthorship Attribution of Small Messages Through Language Models
Theóphilo, A., & Rocha, A. D. R. (2022) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs55849.2022.9975413 · Cited by 1
“Modern LM usually build their vocabulary by building auxiliary models called tokenizers responsible for learning an efficient token set and splitting the raw sentence into vocabulary elements (tokenization) [6, 7].”
BackgroundMethodImplications of Tokenizers in BERT Model for Low-Resource Indian Language
Venkatesan, N., & Arulanand, N. (2023) · Journal of Soft Computing Paradigm · doi:10.36548/jscp.2022.4.005 · Cited by 2
Morphologically-Aware Vocabulary Reduction of Word Embeddings
Chia, C. C., Tkachenko, M., & Lauw, H. W. (2022) · IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT) · doi:10.1109/wi-iat55865.2022.00018 · Cited by 1
“WordPiece [28], BPE [29] and SentencePiece [30] generate vocabulary of a user-specified size and could represent OOV words as segments of the learnt pieces.”
BackgroundPrefix-Level Detection and Autocorrection of Keyboard Input Errors
Bellegarda, J. R. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10096527 · Cited by 1
“Recent advances, however, have resulted in neural solutions based on smaller units, such as single characters and/or variable-length sub-words (referred to as “sentence pieces” or fragments) [10].”
BackgroundLego-Features: Exporting Modular Encoder Features for Streaming and Deliberation ASR
Botros, R., Prabhavalkar, R., Schalkwyk, J., Chelba, C. I., Sainath, T. N., & Beaufays, F. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095464 · Cited by 1
“Our models work with 4,096 word pieces [27].”
MethodSubword Tokenization of Noisy Housing Defect Complaints for Named Entity Recognition
Jeon, K., & Lee, G. (2023) · International Symposium on Automation and Robotics in Construction (ISARC) · doi:10.22260/isarc2023/0057 · Cited by 1
Mensah, S., Sun, K., & Αλέτρας, Ν. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-short.86 · Cited by 1
Event-Based Knowledge MLM for Arabic Event Detection
Yamani, A. Z., Alsulami, A., & Al-Zaidy, R. (2022) · Arabic Natural Language Processing Workshop (WANLP) · doi:10.18653/v1/2022.wanlp-1.25 · Cited by 1
“A WordPiece (Schuster and Nakajima, 2012) tokenizer is trained on the entire dataset (10GB text) with a vocabulary size of 30522 using Hugging”
MethodCan Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?
Li, X., Liu, M., & Gao, S. B. (2023) · Joint Conference on Lexical and Computational Semantics · doi:10.18653/v1/2023.starsem-1.15 · Cited by 1
Analyzing the Impact of Tokenization on Multilingual Epidemic Surveillance in Low-Resource Languages
Mutuvi, S., Boroş, E., Doucet, A., Lejeune, G., Jatowt, A., & Odeo, M. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-41682-8_2 · Cited by 3
Lee, S., & Bloem, J. (2023) · Workshop on Slavic Natural Language Processing (SlavicNLP) · doi:10.18653/v1/2023.bsnlp-1.7 · Cited by 1
“Both models use WordPiece subword tok-enization (Schuster and Nakajima, 2012).”
Leveraging Semantic Representations via Knowledge Graph Embeddings
Krause, F., Kurniawan, K., Kiesling, E., Martínez-Gil, J., Hoch, T., Pichler, M., Heinzl, B., & Moser, B. (2023) · Springer eBooks · doi:10.1007/978-3-031-46452-2_5 · Cited by 3
Efficient Log Anomaly Detection Based on Dimension Reduction and Attention Aware TCN
Xu, Z., Shi, Y., Su, Z., Song, L., Zhang, J., Wang, X., & Li, H. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-2303-4_33 · Cited by 1
Building a Large Corpus and Pre-trained Language Models from National and Local Assembly Minutes
Nagafuchi, K., Kimura, Y., Kadowaki, K., & Araki, K. (2024) · Journal of Natural Language Processing · doi:10.5715/jnlp.31.707 · Cited by 2
Wu, Y., Li, Y., Wang, Z., Tan, Q., Liu, J., & Jiang, Y. (2024) · International Journal of Software Engineering and Knowledge Engineering · doi:10.1142/s021819402450044x · Cited by 1
Towards Understanding Evolution of Science through Language Model Series
Dong, J., Lyu, Z., & Ke, Q. (2026) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3821210 · Cited by 1
“[37] developed a classifier to label citation functions ( e.g. , background, motivation, etc.) and applied it to NLP papers to reveal the evolution of this field.”
LabelEase: A Semi-Automatic Tool for Efficient and Accurate Trace Labeling in Microservices
Zhang, S., Che, Z., Pan, Z., Nie, X., Sun, Y., Pan, L., & Pei, D. (2024) · Proceedings/Proceedings - International Symposium on Software Reliability Engineering · doi:10.1109/issre62328.2024.00032 · Cited by 2
“For out-of-vocabulary (OOV) or new compound words, we employ WordPiece tokenization [24], which involves breaking down words into individual characters and iteratively identifying the most common character combinations until the desired vocabulary is achieved.”
Baeza-Yates, R. A., Murgai, L., & Dudy, S. (2024) · ACM eBooks · doi:10.1145/3674127.3674138 · Cited by 1
Large Language Models for Dummies
Calamo, M., & Marinacci, M. (2025) · Springer eBooks · doi:10.1007/978-3-031-92285-5_3 · Cited by 1
Monakhov, S., Diessel, H., & Balthes, B. (2025) · Cognition · doi:10.1016/j.cognition.2025.106292 · Cited by 1
Security Requirements Classification by Means of Explainable Transformer Models
Petrillo, L., Martinelli, F., Santone, A., & Mercaldo, F. (2025) · IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) · doi:10.1109/eurospw67616.2025.00064 · Cited by 1
“The first is WordPiece [16], a combination of word-level and subword-level tokenization, to divide words into subwords or word pieces depending on the frequency of the subwords in the training data.”
Optimally Segmenting Inputs for NMT Shows Preference for Character-Level ProcessingPreprint
Kreutzer, J., & Sokolov, A. (2018) · arXiv (Cornell University) · Cited by 1
Krishna, A., Santra, B., Bandaru, S. P., Sahu, G., Sharma, V. D., Satuluri, P., & Goyal, P. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.01446 · Cited by 2
“cal tagging is not straightforward, as they learn a new sub-word vocabulary using the sentencepiece model (Schuster and Nakajima, 2012).”
BackgroundDeliberation Model Based Two-Pass End-to-End Speech RecognitionPreprint
Hu, K., Sainath, T. N., Pang, R., & Prabhavalkar, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.07962 · Cited by 1
“The LAS decoder has a 4,096-dimensional softmax layer to predict the same mixed-case wordpieces [19] as the RNN-T.”
BackgroundMethodInducing Meaningful Units from Character Sequences with Slot Attention.Preprint
Behjati, M., & Henderson, J. (2021) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 2
Improving BERT Pretraining with Syntactic SupervisionPreprint
Tziafas, G., Kogkalidis, K., Wijnholds, G., & Moortgat, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.10516 · Cited by 1
“We tokenize words using a preconstructed WordPiece (Schuster and Naka-jima, 2012) vocabulary of 30 000 tokens based on a larger collection of written Dutch corpora (Vries et al., 2019).”
MethodBalakrishnan, R., Sloan, S., & Aswani, A. J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.09495 · Cited by 1
“Because of the importance of n-grams in author identification, we also employ Wordpiece Tokenization (Schuster and Naka-jima 2012), which breaks down a word into the most common subword units and ensures that every word can be tok-enized.”
Wine is Not v i n. -- On the Compatibility of Tokenizations Across LanguagesPreprint
Moens (Ed.), M.-F., Huang (Ed.), X., Specia (Ed.), L., & Yih (Ed.), S. W.-T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.05772 · Cited by 1
“Due to memory constraints and to overcome sparsity, subword tokenization models such as byte pair encoding (Sennrich et al., 2016) or WordPiece tokenization (Schuster and Nakajima, 2012) are most commonly used.”
BackgroundMethodInfluential(2021) · PeerJ Computer Science · doi:10.7287/peerj-cs.443v0.1/reviews/2 · Cited by 1
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.004 · Cited by 1
Rethinking Data-driven Networking with Foundation Models: Challenges and OpportunitiesPreprint
Le, F., Srivatsa, M., Ganti, R., & Sekar, V. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2211.06494 · Cited by 2
“[68] Mike Schuster and Kaisuke Nakajima.”
MethodInfluentialKnafou, J., Haas, Q., Borissov, N., Counotte, M. J., Low, N. N., Imeri, H., Ipekci, A. M., Buitrago‐García, D., Heron, L., Amini, P., et al. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.01.18.524571 · Cited by 1
A Cohesive Distillation Architecture for Neural Language ModelsPreprint
Wahle, J. P. (2023) · Authorea · doi:10.22541/au.167528147.79728645/v1 · Cited by 1
Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?Preprint
Li, X., Liu, M., & Gao, S. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.15268 · Cited by 1
“, 2019) accepts inputs from a Wordpiece tokenizer (Schuster and Nakajima, 2012), while RoBERTa (Liu et al.”
MethodProteasomal cleavage prediction: state-of-the-art and future directionsPreprint
Ziegler, I., Ma, B., Bischl, B., Dorigatti, E., & Schubert, B. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.07.17.549305 · Cited by 1
“[35] Mike Schuster and Kaisuke Nakajima.”
MethodNeural Models for Source Code Synthesis and CompletionPreprint
Niyogi, M. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.06690 · Cited by 2
Multilingual Tokenization Efficiency in Large Language Models: A Study on Indian LanguagesPreprint
Mohamed, A., & Dhamodharan, B. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5201887 · Cited by 2
Log analysis and anomaly detection in log files with natural language processing techniques
Dhakal, K. (2023) · Aaltodoc (Aalto University) · Cited by 1
Kuuluvainen, E. (2023) · Theseus (Ammattikorkeakoulujen) · Cited by 1
Zhang, S., Xu, A., Chen, Q., Zhao, H., Cui, W., Zheng, N., Lin, H., Liu, X., & Guo, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.09560 · Cited by 1
“It is beneficial for each request to discard its private consistent context, and look for the most timely context to generate responses.”
Automatic sentence simplification using controllable and unsupervised methods
Martin, L. (2021) · theses.fr (ABES) · doi:10.70675/e997dea7zc65ez46ecz924dzc86bd7869f1c · Cited by 9
Deep learning for information extraction from business documents
Sage, C. (2021) · theses.fr (ABES) · doi:10.70675/fd1f6c1fz859bz4230zbf7bz9bb7cdb83ad0 · Cited by 3
Epidemic Event Extraction in Multilingual and Low-resource Settings
Mutuvi, S. (2022) · theses.fr (ABES) · doi:10.70675/82e91446zf499z4438zaaa7z66548ffb98b0 · Cited by 3
Leveraging linguistic and semantic information for relation extraction from domain-specific texts
Tang, A. (2023) · theses.fr (ABES) · doi:10.70675/5e2de3fdz3bb1z4e21z877ez380f3d91a22a · Cited by 3
Where to cut, how deep: BPE and Unigram-LM on chemistry SMILESPreprint
Heidenreich, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.05691 · Cited by 1
“(WordPiece [41] builds bottom-up like BPE, differing only in its merge criterion, and is out of scope.)”
Joint Optimization for Greedy Longest-match TokenizationPreprint
Singh, A., Mody, D., Shdaifat, G. A., Alshamy, H., Wiemerslage, A., Reddy, V., & Schmidt, C. W. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.23362 · Cited by 1
“Byte Pair Encoding (BPE; Sennrich et al., 2016) is among the most frequently used subword tokenizers; WordPiece (Wu et al., 2016; Schuster & Nakajima, 2012) and UnigramLM (Kudo, 2018) are also widely deployed.”
ModernBERT-TR: A Modern Encoder Foundation Model for TurkishPreprint
Alkurdi, B., Kesgin, H. T., Yuce, M. K., & Amasyalı, M. F. (2026) · Research Square · doi:10.21203/rs.3.rs-9500377/v1 · Cited by 1
“Like Ippolito et al. (2020), we only consider texts that have at least 192 WordPiece (Schuster and Nakajima, 2012) tokens.”
MethodSugiura, K., Lee, R., Kashioka, H., Zettsu, K., & Kidawara, Y. (2013) · IEEE International Conference on Mobile Data Management (MDM) · doi:10.1109/mdm.2013.96
“More and more attention from both academics and industry is being paid to voice search [1], [2], speechto-speech translation [3], and voice user interfaces (VUI) for mobile devices [4].”
BackgroundOur Neural Machine Translation Systems for WAT 2019
Yang, W., & Ogata, J. (2019) · Workshop on Asian Translation (WAT) · doi:10.18653/v1/d19-5220
“BPE in “OpenNMT-py” and “wordpieces” (Schuster and Nakajima, 2012; Wu et al., 2016) in “Tensor2Tensor”.”
Character-Aware Attention-Based End-to-End Speech RecognitionPreprint
Meng, Z., Gaur, Y., Li, J., & Gong, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.01795
“One popular type of WSUs is the word pieces model generated by iteratively combining two units out of the current inventory that increase the likelihood the most on the training data [13, 16].”
MethodNeural machine translation for multimodal interaction
Chowdhury, K. D. (2019) · Dublin City University Open Access Institutional Repository (Dublin City University)
EXTENDED CONVOLUTIONAL NEURAL NETWORKS POST-TRAINED WITH FACTORED STATISTICAL MACHINE
Kamgang, V. B., & Nkenlifack, M. (2020) · International Research Journal of Computer Science · doi:10.26562/irjcs.2020.v0707.003
“For the WMT’14 English to German translation task we used the following prior work for comparison: an attention based four layer LSTM model [36] a character without attention based convolutional model proposed by ByteNet [35] set to operate the encoder and decoder with 30 layers each, an encoder-decoder architecture b…”
MethodA Study on Korean Language Model Based on Long Short-Term Memory (LSTM)
Lee, S. (2020) · doi:10.22716/sckt.2020.8.1.003
“한국어의 경우에는 형태 소가 언어처리의 기본으로 인식되고 있으나 최근에는 word piece model (WPM) 등이 많이 사용되고 있다([6]).”
A Study on Korean Language Model Based on Attention
Lee, S. (2020) · The Society of Convergence Knowledge Transactions · doi:10.22716/sckt.2020.8.4.033
“WPM 모델은 음절을 기준으로 하여 통계적인 사용이 많으면 음절 이 확대되는 방법을 취한다.”
InfluentialAI for Healthcare: Diagnosis, Clinical-Trial Matching, and Patient Recruitment
Das, K. A. (2020) · OakTrust (Texas A&M University Libraries)
Evaluating Transferability of BERT Models on Uralic Languages
Ács, J., Lévai, D., & Kornai, A. (2021) · SZTAKI Publication Repository (Hungarian Academy of Sciences) · doi:10.48550/arxiv.2109.06327
“…Livvi [olo]; Komi Permyak [koi]; Komi Zyrian [kpv]; Northern Sámi [sme]; and Skolt Sámi [sms] have UD treebanks (Rueter and Tyers, 2018; Rueter, 2018; Pirinen, 2019; Rueter, 2014; Rueter et al., 2020; Partanen et al., 2018; Sheyanova and Tyers, 2017), but these are considerably smaller in size.”
BackgroundKoehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.015
Text Counterfactuals via Latent Optimization and Shapley-Guided SearchPreprint
Pope, Q., & Fern, X. Z. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.11589 · Cited by 1
“HotFlip implemented in TextAttack only works for word or characterlevel classifiers, not WordPiece (Schuster and Nakajima, 2012) classifiers like RoBERTa and BERT.”
MethodBenchmarking down-scaled (not so large) pre-trained language models
Aßenmacher, M., Schulze, P., & Heumann, C. (2021) · Open access LMU (Ludwid Maxmilian's Universitat Munchen) · doi:10.5282/ubm/epub.77476 · Cited by 1
Clustering Monolingual Vocabularies to Improve Cross-Lingual Generalization
Bassani, R., Søgaard, A., & Deoskar, T. (2021) · Workshop on Multilingual Representation Learning (MRL) · doi:10.18653/v1/2021.mrl-1.3
“Moreover, mBERT relies on a single, multilingual tokenizer to create its vocabulary (WordPiece, Schuster and Nakajima (2012)), a subword tok-enizer based on an algorithm similar to BPE (Gage, 1994).”
MethodAnalysis of Word Dependency Relations and Subword Models in Abstractive Text Summarization
Ozkan, A., & Güngör, T. (2021) · Innovations in Intelligent Systems and Applications Conference (ASYU) · doi:10.1109/asyu52992.2021.9599041
“We used byte-pair encoding [9], WordPiece [10] and un-igram language model [11] as the subword models.”
MethodCORPUS PARA O DOMÍNIO ACADÊMICO: MODELOS E APLICAÇÕES
Pinto, I. D. J. P. (2021) · doi:10.17771/pucrio.acad.55901
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.016
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.013
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.001
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.007
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.010
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.020
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.011
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.012
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.006
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.014
(2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.019
ANALIZA RAZLIČITIH MODELA ZA PREPOZNAVANJE IMENOVANIH ENTITETA NA SRPSKOM JEZIKU
Cvejić, A. (2022) · Zbornik radova Fakulteta tehničkih nauka u Novom Sadu · doi:10.24867/16be40cvejic
Koehn, P. (2020) · Cambridge University Press eBooks · doi:10.1017/9781108608480.021
Neural machine translation with limited resources
Mohiuddin, T. (2022) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/157475
Avram, C., & Rusu, R. (2021) · Risk in Contemporary Economy · doi:10.35219/rce20670532134
Blanchard, A., Zhang, P., Bhowmik, D., Mehta, K. R., Gounley, J. P., Reeve, S. T., Irle, S., & Pasini, M. L. (2022) · ChemRxiv · doi:10.26434/chemrxiv-2022-gw1n3-v2
A Survey of Machine Translation Technologies for Machine Translation on Patent Documents
Imamura, K., Echizen‐ya, H., Ehara, T., Goto, I., Sudoh, K., Sonoh, S., Tsunakawa, T., Nakazawa, T., Ninomiya, T., & Wang, X. (2022) · Journal of Natural Language Processing · doi:10.5715/jnlp.29.925 · Cited by 1
Knowledge Graph Completion Method Combined With Adaptive Enhanced Semantic InformationPreprint
Yin, Z., Ji, W., Zhou, G., Yue, Y., Zhang, X., & Sun, C. (2022) · doi:10.22541/au.166741826.68495376/v1
Language Resource Construction of Multi-Domain Philippine English Text for Pre-training Objective
Canon, M. J. P., Sy, C. Y., Palaoag, T. D., Roxas, R. E., & Maceda, L. L. (2022) · International Conference on Advanced Computer Science and Information Systems · doi:10.1109/icacsis56558.2022.9923429
“The algorithm was outlined in Japanese and Korean Voice Search [30] and was benchmarked by Wu et.al [31].”
MethodAbstractive Turkish Text Summarization and Cross-Lingual Summarization Using Transformer
Taspinar, E. K., Yetis, Y. B., & Cihan, O. (2022) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-6684-6001-6.ch011
NEEBS: Nonexpert large‐scale environment building system for deep neural network
Tajima, Y., Asaoka, M., Tabuchi, A., Kasagi, A., & Tabaru, T. (2022) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.7499
Text Classification Models and Topic Models: An Overall Picture and a Case Study in Vietnamese
Lâm, K. N., Tran, V.-L. L., & Kalita, J. K. (2022) · Communications in computer and information science · doi:10.1007/978-981-19-8069-5_25
Blanchard, A., Zhang, P., Bhowmik, D., Mehta, K. R., Gounley, J. P., Reeve, S. T., Irle, S., & Pasini, M. L. (2025) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)
Decomposition of construction method for a language encoder
Трофимов, И. В. (2023) · Program systems theory and applications · doi:10.25209/2079-3316-2023-14-1-31-54
A Bag of Tokens Neural Network to Predict Webpage Age
Meinke, K., der Laan, T. A. V., Iancu, T., & Cakir, C. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-34671-2_12
Yehoshyna, H., O., Voronoy, S. M., Polıkarovskykh, O., & Gokhman, R. (2023) · Scientific papers of Donetsk National Technical University Series Informatics Cybernetics and Computer Science · doi:10.31474/1996-1588-2023-1-36-51-60
Clemson NLP at SemEval-2023 Task 7: Applying GatorTron to Multi-Evidence Clinical NLI
Alameldin, A., & Williamson, A. (2023) · International Workshop on Semantic Evaluation (SemEval) · doi:10.18653/v1/2023.semeval-1.220 · Cited by 4
“We tokenize the datasets using WordPiece tokenization (Schuster and Nakajima, 2012) with the 50,000 token clinical vocabulary released alongside the GatorTron model 2 (Yang et al., 2022).”
MethodCrisostomi, D., Manzotti, A., Palumbo, E., Bernardi, D., Campbell, S., & Garg, S. (2022) · doi:10.18653/v1/2022.mmnlu-1.5
“Since the tokenizer is based on the Wordpiece model (Schuster and Nakajima, 2012), words are often split in subwords: the word ‘microfono’ for example would be split in ‘micro’ and ‘##fono’.”
MethodAutomatic programming (source code generator) based on an ontological model
David, P. E., Malathi, S., & Anandhakumar, P. (2023) · Advances in computers · doi:10.1016/bs.adcom.2023.08.008
de Souza, F. C. (2020) · doi:10.47749/t/unicamp.2020.1161906
Metadata
- Authors
- Schuster, M., & Nakajima, K.
- Affiliations
- Google (United States) (US)
- Venue
- 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2012-03
- Pages
- 5149-5152
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W2121879602
- Cited by
- 862
- References
- 8
- Citations by year
- 2026: 36 · 2025: 66 · 2024: 90 · 2023: 124 · 2022: 137 · 2021: 155 · 2020: 144 · 2019: 59 · 2018: 36 · 2017: 10 · 2016: 3 · 2013: 2
- FWCI
- 0.49
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- voice search, Japanese language, Korean language, acoustic modeling
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.
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.