Filters








1,523 Hits in 5.3 sec

Spoken Language Communication with Machines: The Long and Winding Road from Research to Business [chapter]

Roberto Pieraccini, David Lubensky
2005 Lecture Notes in Computer Science  
This paper traces the history of spoken language communication with computers, from the first attempts in the 1950s, through the establishment of the theoretical foundations in the 1980s, to the incremental  ...  Then a perspective is given on the current conversational technology market and industry, with an analysis of its business value and commercial models.  ...  However, even in those situations, a higher automation rate can be obtained by using speech technology and by a complete redesign of the user interface.  ... 
doi:10.1007/11504894_2 fatcat:3akuel7fdzd4xo334mh7zjgyai

Bootstrapping a Crosslingual Semantic Parser [article]

Tom Sherborne, Yumo Xu, Mirella Lapata
2020 arXiv   pre-print
We develop a Transformer-based parser combining paraphrases by ensembling attention over multiple encoders and present new versions of ATIS and Overnight in German and Chinese for evaluation.  ...  Considering when MT is inadequate, we also find that using our approach achieves parsing accuracy within 2% of complete translation using only 50% of training data.  ...  Acknowledgements The authors gratefully acknowledge the support of the UK Engineering and Physical Sciences Research Council (grant EP/L016427/1; Sherborne) and the European Research Council (award number  ... 
arXiv:2004.02585v4 fatcat:pctm3axvcfd5bcg36doqa3a46q

Zero-Shot Visual Slot Filling as Question Answering [article]

Larry Heck, Simon Heck
2022 arXiv   pre-print
This paper presents a new approach to slot filling by reformulating the slot filling task as Question Answering, and replacing slot tags with rich natural language questions that capture the semantics  ...  The multi-task approach facilitates the incorporation of a large number of successive refinements and transfer learning across tasks.  ...  A primary component of the PDAs is Natural Language Understanding (NLU) -understanding the meaning of the user's utterance.  ... 
arXiv:2011.12340v2 fatcat:wzfvi57j5bedzlmm4qhbpx2oba

Comparative study on the customization of natural language interfaces to databases

Rodolfo A. Pazos R., Marco A. Aguirre L., Juan J. González B., José A. Martínez F., Joaquín Pérez O., Andrés A. Verástegui O.
2016 SpringerPlus  
The purpose of these systems is to facilitate the querying task to users by sparing them the burden of having to learn a DB query Abstract In the last decades the popularity of natural language interfaces  ...  NLIDBs are important tools since they allow users accessing information in a database by a query formulated in natural language (NL).  ...  With a NLIDB, a user simply types a query in natural language, similarly as he/ she would do when communicating with another person, and the interface interprets the query and translates it to a DB query  ... 
doi:10.1186/s40064-016-2164-y pmid:27190752 pmcid:PMC4851672 fatcat:dukc2w2zaffgxnris5odmxy2q4

An Integrated Model-Driven Approach for Mechatronic Systems Testing

Roberto S. Silva Filho, Christof J. Budnik
2012 2012 IEEE Fifth International Conference on Software Testing, Verification and Validation  
The result is unprecedented levels of automation and testability. We discuss the design and implementation of our approach, illustrating its use with a hybrid calculator example.  ...  As such, the adequate testing and validation of such infrastructures are of prime importance.  ...  SPECIFICATION LANGUAGE In this section, we describe key elements of the specification language used in the system, and the code generation transformations.  ... 
doi:10.1109/icst.2012.125 dblp:conf/icst/FilhoB12 fatcat:ybpzol5jejgovjqk6hn7wgnjgm

Unsupervised Transfer Learning for Spoken Language Understanding in Intelligent Agents [article]

Aditya Siddhant, Anuj Goyal, Angeliki Metallinou
2018 arXiv   pre-print
The improvements are more pronounced in low resource settings and when using only 1000 labeled in-domain samples, our techniques match the performance of training from scratch on 10-15x more labeled in-domain  ...  In this paper, we explore techniques to efficiently transfer the knowledge from these unlabeled utterances to improve model performance on Spoken Language Understanding (SLU) tasks.  ...  ELMoL+ST This is also a three step method where in the first step, the lower layer of multi-task IC/ET architecture is pre-trained using a language model objective.  ... 
arXiv:1811.05370v1 fatcat:ss646d5c5vfmvdt3tjd2rquvke

Unsupervised Transfer Learning for Spoken Language Understanding in Intelligent Agents

Aditya Siddhant, Anuj Goyal, Angeliki Metallinou
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
The improvements are more pronounced in low resource settings and when using only 1000 labeled in-domain samples, our techniques match the performance of training from scratch on 10-15x more labeled in-domain  ...  In this paper, we explore techniques to efficiently transfer the knowledge from these unlabeled utterances to improve model performance on Spoken Language Understanding (SLU) tasks.  ...  ELMoL+ST This is also a three step method where in the first step, the lower layer of multi-task IC/ET architecture is pre-trained using a language model objective.  ... 
doi:10.1609/aaai.v33i01.33014959 fatcat:ttcydusdfngengs72ppbtyalle

NLP commercialisation in the last 25 years

Robert Dale
2019 Natural Language Engineering  
AbstractThe Journal of Natural Language Engineering is now in its 25th year.  ...  The editorial preface to the first issue emphasised that the focus of the journal was to be on the practical application of natural language processing (NLP) technologies: the time was ripe for a serious  ...  Dialog interfaces: Interactive interfaces that permit a user to ask a natural language question (either via speech or text) and receive an answer, possibly in the context of a longer multi-turn dialog.  ... 
doi:10.1017/s1351324919000135 fatcat:7bfrwfaxwvaolcqv5snvl46dne

An object-oriented relational database

William J. Premerlani, James E. Rumbaugh, Michael R. Blaha, Thomas A. Varwig
1990 Communications of the ACM  
(OMG), and A Tool Integration System (ATIS).  ...  in underling heterogeneity -The user interface hides the heterogeneous nature of the system by providing the run-time mechanisms for implementing the client-server computation model. "* Browsers -KBE-SLCSE  ...  A task then needs to be performed in selecting commercial tools and integrating their control and data in terms of the defined services. "* Defining a process model which allows the description of alternate  ... 
doi:10.1145/92755.92772 fatcat:dsay3vjnl5a63cfpczwclrb5iu

CoSQL: A Conversational Text-to-SQL Challenge Towards Cross-Domain Natural Language Interfaces to Databases [article]

Tao Yu, Rui Zhang, He Yang Er, Suyi Li, Eric Xue, Bo Pang, Xi Victoria Lin, Yi Chern Tan, Tianze Shi, Zihan Li, Youxuan Jiang, Michihiro Yasunaga, Sungrok Shim (+9 others)
2019 arXiv   pre-print
CoSQL introduces new challenges compared to existing task-oriented dialogue datasets:(1) the dialogue states are grounded in SQL, a domain-independent executable representation, instead of domain-specific  ...  When user questions are answerable by SQL, the expert describes the SQL and execution results to the user, hence maintaining a natural interaction flow.  ...  CONFORM SQL We convert SQL written by us back to natural language. (We should use the column names and values in the SQL).  ... 
arXiv:1909.05378v1 fatcat:6acielnwvvaptov5ipmnddrkwi

When BERT Meets Quantum Temporal Convolution Learning for Text Classification in Heterogeneous Computing [article]

Chao-Han Huck Yang, Jun Qi, Samuel Yen-Chi Chen, Yu Tsao, Pin-Yu Chen
2022 arXiv   pre-print
Our experiments on intent classification show that our proposed BERT-QTC model attains competitive experimental results in the Snips and ATIS spoken language datasets.  ...  Particularly, the BERT-QTC boosts the performance of the existing quantum circuit-based language model in two text classification datasets by 1.57% and 1.52% relative improvements.  ...  BERTs [27] can be used for a wide variety of language tasks since different SLU tasks only require a few more layers to the core model for finetuning.  ... 
arXiv:2203.03550v1 fatcat:dsn5jjemnvcy5owb5zvqt3tvem

Session 2: DARPA Resource Management and ATIS Benchmark Test Poster Session

David S. Pallett
1991 North American Chapter of the Association for Computational Linguistics  
All papers in this poster session at the DARPA speech workshop reported results obtained using the Benchmark Test Material.  ...  The Workshop Planning Committee suggested a three-part session format consisting of: (1) Introductory remarks, (2) one hour to review and discuss posters, and (3) open discussion.  ...  of a Commercial Natural Language Interface in the ATIS Task" [13] .  ... 
dblp:conf/naacl/Pallett91 fatcat:2bsf23s6g5fllcjd6cldq2l7ia

The role of voice input for human-machine communication

P. R. Cohen, S. L. Oviatt
1995 Proceedings of the National Academy of Sciences of the United States of America  
Optimism is growing that the near future will witness rapid growth in human-computer interaction using voice.  ...  to the successful development and commercialization of spoken language systems.  ...  Natural Language In general, because the human-machine communication in spoken language involves the system's understanding a natural language, but not the entire language, users will employ constructs  ... 
doi:10.1073/pnas.92.22.9921 pmid:7479803 pmcid:PMC40712 fatcat:js3j6ovthzcujhfhumwke5tqfi

Knowledge Distillation from BERT Transformer to Speech Transformer for Intent Classification [article]

Yidi Jiang, Bidisha Sharma, Maulik Madhavi, Haizhou Li
2021 arXiv   pre-print
In this work, we exploit the scope of the transformer distillation method that is specifically designed for knowledge distillation from a transformer based language model to a transformer based speech  ...  End-to-end intent classification using speech has numerous advantages compared to the conventional pipeline approach using automatic speech recognition (ASR), followed by natural language processing modules  ...  There is a wide range of interactive interface-based applications requiring SLU systems, namely, voice search and chatbot on mobile devices, attracting interest from both commercial and academic sectors  ... 
arXiv:2108.02598v1 fatcat:5zmqe5wxljc6npa77uf5dsithy

CoSQL: A Conversational Text-to-SQL Challenge Towards Cross-Domain Natural Language Interfaces to Databases

Tao Yu, Rui Zhang, Heyang Er, Suyi Li, Eric Xue, Bo Pang, Xi Victoria Lin, Yi Chern Tan, Tianze Shi, Zihan Li, Youxuan Jiang, Michihiro Yasunaga (+12 others)
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
CoSQL introduces new challenges compared to existing task-oriented dialogue datasets: (1) the dialogue states are grounded in SQL, a domain-independent executable representation, instead of domain-specific  ...  When user questions are answerable by SQL, the expert describes the SQL and execution results to the user, hence maintaining a natural interaction flow.  ...  This work was supported in part by IBM under the Sapphire Project at the University of Michigan.  ... 
doi:10.18653/v1/d19-1204 dblp:conf/emnlp/YuZELXPLTSLJYSC19 fatcat:k2hth3eegfaerkb2hycu4tehgi
« Previous Showing results 1 — 15 out of 1,523 results