{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/joint-online-spoken-language-understanding","title":"Joint Online Spoken Language Understanding and Language Modeling with Recurrent Neural Networks","arxiv_id":"1609.01462","date":"2016-09-06","proceeding":"WS 2016 9","authors":["Bing Liu","Ian Lane"],"abstract":"Speaker intent detection and semantic slot filling are two critical tasks in\nspoken language understanding (SLU) for dialogue systems. In this paper, we\ndescribe a recurrent neural network (RNN) model that jointly performs intent\ndetection, slot filling, and language modeling. The neural network model keeps\nupdating the intent estimation as word in the transcribed utterance arrives and\nuses it as contextual features in the joint model. Evaluation of the language\nmodel and online SLU model is made on the ATIS benchmarking data set. On\nlanguage modeling task, our joint model achieves 11.8% relative reduction on\nperplexity comparing to the independent training language model. On SLU tasks,\nour joint model outperforms the independent task training model by 22.3% on\nintent detection error rate, with slight degradation on slot filling F1 score.\nThe joint model also shows advantageous performance in the realistic ASR\nsettings with noisy speech input.","url_abs":"http://arxiv.org/abs/1609.01462v1","url_pdf":"http://arxiv.org/pdf/1609.01462v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"intent-detection","task_name":"Intent Detection"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"slot-filling","task_name":"Slot Filling"},{"task_slug":"spoken-language-understanding","task_name":"Spoken Language Understanding"},{"task_slug":"slot-filling-1","task_name":"slot-filling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/intent-detection-on-atis","task":"Intent Detection","dataset":"ATIS","model":"Joint model with recurrent slot label context","rank_in_archive_order":4,"of":16,"metrics":{"Accuracy":"98.40"},"uses_additional_data":false},{"leaderboard":"/sota/slot-filling-on-atis","task":"Slot Filling","dataset":"ATIS","model":"Joint model with recurrent slot label context","rank_in_archive_order":14,"of":14,"metrics":{"F1":"0.9464"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1609.01462","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}