{"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/syllable-based-sequence-to-sequence-speech","title":"Syllable-Based Sequence-to-Sequence Speech Recognition with the Transformer in Mandarin Chinese","arxiv_id":"1804.10752","date":"2018-04-28","proceeding":null,"authors":["Shiyu Zhou","Linhao Dong","Shuang Xu","Bo Xu"],"abstract":"Sequence-to-sequence attention-based models have recently shown very\npromising results on automatic speech recognition (ASR) tasks, which integrate\nan acoustic, pronunciation and language model into a single neural network. In\nthese models, the Transformer, a new sequence-to-sequence attention-based model\nrelying entirely on self-attention without using RNNs or convolutions, achieves\na new single-model state-of-the-art BLEU on neural machine translation (NMT)\ntasks. Since the outstanding performance of the Transformer, we extend it to\nspeech and concentrate on it as the basic architecture of sequence-to-sequence\nattention-based model on Mandarin Chinese ASR tasks. Furthermore, we\ninvestigate a comparison between syllable based model and context-independent\nphoneme (CI-phoneme) based model with the Transformer in Mandarin Chinese.\nAdditionally, a greedy cascading decoder with the Transformer is proposed for\nmapping CI-phoneme sequences and syllable sequences into word sequences.\nExperiments on HKUST datasets demonstrate that syllable based model with the\nTransformer performs better than CI-phoneme based counterpart, and achieves a\ncharacter error rate (CER) of \\emph{$28.77\\%$}, which is competitive to the\nstate-of-the-art CER of $28.0\\%$ by the joint CTC-attention based\nencoder-decoder network.","url_abs":"http://arxiv.org/abs/1804.10752v2","url_pdf":"http://arxiv.org/pdf/1804.10752v2.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":"syllable-based-sequence-to-sequence-speech","repo_url":"https://github.com/gentaiscool/end2end-asr-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"sequence-to-sequence-speech-recognition","task_name":"Sequence-To-Sequence Speech Recognition"},{"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":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.10752","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}