{"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/returnn-as-a-generic-flexible-neural-toolkit","title":"RETURNN as a Generic Flexible Neural Toolkit with Application to Translation and Speech Recognition","arxiv_id":"1805.05225","date":"2018-05-14","proceeding":"ACL 2018 7","authors":["Albert Zeyer","Tamer Alkhouli","Hermann Ney"],"abstract":"We compare the fast training and decoding speed of RETURNN of attention\nmodels for translation, due to fast CUDA LSTM kernels, and a fast pure\nTensorFlow beam search decoder. We show that a layer-wise pretraining scheme\nfor recurrent attention models gives over 1% BLEU improvement absolute and it\nallows to train deeper recurrent encoder networks. Promising preliminary\nresults on max. expected BLEU training are presented. We are able to train\nstate-of-the-art models for translation and end-to-end models for speech\nrecognition and show results on WMT 2017 and Switchboard. The flexibility of\nRETURNN allows a fast research feedback loop to experiment with alternative\narchitectures, and its generality allows to use it on a wide range of\napplications.","url_abs":"http://arxiv.org/abs/1805.05225v2","url_pdf":"http://arxiv.org/pdf/1805.05225v2.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":[{"paper_slug":"returnn-as-a-generic-flexible-neural-toolkit","repo_url":"https://github.com/rwth-i6/returnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"returnn-as-a-generic-flexible-neural-toolkit","repo_url":"https://github.com/danenergetics/returnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"returnn-as-a-generic-flexible-neural-toolkit","repo_url":"https://github.com/papar22/returnn-my-branch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.05225","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}