{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T17:25:40Z","timestamp":1777656340598,"version":"3.51.4"},"reference-count":23,"publisher":"IEEE","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2017,10]]},"DOI":"10.1109\/waspaa.2017.8170058","type":"proceedings-article","created":{"date-parts":[[2017,12,13]],"date-time":"2017-12-13T14:38:44Z","timestamp":1513175924000},"page":"374-378","source":"Crossref","is-referenced-by-count":72,"title":["Automated audio captioning with recurrent neural networks"],"prefix":"10.1109","author":[{"given":"Konstantinos","family":"Drossos","sequence":"first","affiliation":[]},{"given":"Sharath","family":"Adavanne","sequence":"additional","affiliation":[]},{"given":"Tuomas","family":"Virtanen","sequence":"additional","affiliation":[]}],"member":"263","reference":[{"key":"ref10","article-title":"Google's neural machine translation system: Bridging the gap between human and machine translation","author":"wu","year":"2016","journal-title":"arXiv 1609 08144"},{"key":"ref11","article-title":"Google's multilingual neural machine translation system: Enabling zero-shot translation","author":"johnson","year":"2016","journal-title":"arXiv 1611 04558"},{"key":"ref12","article-title":"Maxout networks","author":"goodfellow","year":"2013","journal-title":"International Conference on Machine Learning (ICML)"},{"key":"ref13","article-title":"Microsoft COCO captions: Data collection and evaluation server","volume":"abs 1504 325","author":"chen","year":"2015","journal-title":"CoRR"},{"key":"ref14","first-page":"311","article-title":"BLEU: A method for automatic evaluation of machine translation","author":"papineni","year":"2002","journal-title":"Proc 40th Ann Meeting Assoc for Computational Linguistics (ACL 02)"},{"key":"ref15","first-page":"25","article-title":"Rouge: a package for automatic evaluation of summaries","author":"lin","year":"2004","journal-title":"Proceedings of the Workshop on Text Summarization Branches Out (WAS 2004)"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.3115\/1626355.1626389"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7299087"},{"key":"ref18","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2015","journal-title":"International Conference on Learning Representations"},{"key":"ref19","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"glorot","year":"2010","journal-title":"Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS10)"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298935"},{"key":"ref3","year":"2015","journal-title":"Master Library 2 0"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/N15-1173"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.503"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2016.7532983"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2010.5495973"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952260"},{"key":"ref9","article-title":"Neural machine translation by jointly learning to align and translate","author":"bahdanau","year":"2015","journal-title":"International Conference on Learning Representations (ICLR)"},{"key":"ref20","author":"chollet","year":"2015","journal-title":"Keras"},{"key":"ref22","author":"kelly","year":"2016","journal-title":"PyEnchant"},{"key":"ref21","article-title":"Theano: A Python framework for fast computation of mathematical expressions","volume":"abs 1605 2688","year":"2016","journal-title":"ArXiv e-prints"},{"key":"ref23","author":"atkinson","year":"2016","journal-title":"GNU Aspell"}],"event":{"name":"2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)","location":"New Paltz, NY","start":{"date-parts":[[2017,10,15]]},"end":{"date-parts":[[2017,10,18]]}},"container-title":["2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8125199\/8169982\/08170058.pdf?arnumber=8170058","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2018,1,17]],"date-time":"2018-01-17T18:13:30Z","timestamp":1516212810000},"score":1,"resource":{"primary":{"URL":"http:\/\/ieeexplore.ieee.org\/document\/8170058\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,10]]},"references-count":23,"URL":"https:\/\/doi.org\/10.1109\/waspaa.2017.8170058","relation":{},"subject":[],"published":{"date-parts":[[2017,10]]}}}