{"arxiv_id":"1511.04868","incoming":{"papers":20,"sentences":27,"groups":[{"kind":"background","label":"Cites it as background","papers":10,"sentences":13,"quotes":[{"arxiv_id":"2103.00422","title":"Alignment Knowledge Distillation for Online Streaming Attention-based Speech Recognition","url":"/paper/2103.00422","section":"Introduction","sentence":"To make AED models streamable, various methods have been proposed: neural transducer (NT) [ref] , triggered attention [ref] , adaptive computation steps (ACS) [ref] , continuous integrate-and-fire (CIF) [ref] , streaming chunk-aware multi-head attention (SCAMA) [ref] , local windowing [ref] , Gaussian mixture model (GMM) attention [ref] , neural autoregressive transducer (NAT) [ref] , and monotonic chunkwise attention (MoChA) [ref] .","intent_p":0.95,"contests_p":null},{"arxiv_id":"2004.05009","title":"Minimum Latency Training Strategies for Streaming Sequence-to-Sequence ASR","url":"/paper/2004.05009","section":"Introduction","sentence":"To address this, several novel streaming attention-based S2S models have been proposed including local windowing methods [ref] , neural transducer [ref] , hard monotonic attention [ref] , adaptive computation steps [ref] , triggered attention [ref] , and continuous integrate-and-fire [ref] .","intent_p":0.9,"contests_p":null},{"arxiv_id":"1706.06428","title":"An online sequence-to-sequence model for noisy speech recognition","url":"/paper/1706.06428","section":"Relation To Prior Work","sentence":"Finally, Jaitly et al. [ref] used an online sequence-to-sequence method with conditioning on partial inputs, which yielded encouraging results on the TIMIT dataset.","intent_p":0.95,"contests_p":null},{"arxiv_id":"2002.02562","title":"Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss","url":"/paper/2002.02562","section":"Introduction","sentence":"Streamable end-to-end modeling architectures such as the Recurrent Neural Network Transducer (RNN-T) [ref] , Recurrent Neural Aligner (RNA) [ref] , and Neural Transducer [ref] utilize an encoder-decoder based framework where both encoder and decoder are layers of RNNs that generate features from audio and labels respectively.","intent_p":0.9,"contests_p":null},{"arxiv_id":"2005.10113","title":"A Comparison of Label-Synchronous and Frame-Synchronous End-to-End Models for Speech Recognition","url":"/paper/2005.10113","section":"Introduction","sentence":"The representative models in them are originally proposed for neural machine translation (NMT) [ref] , and rely on the attention mechanism to extract relevant speech information from the encoded frames, such as the attention-based models [ref] and transformer [ref] .","intent_p":0.86,"contests_p":null}]}],"unlabelled_sentences":14},"outgoing":{"papers":7,"sentences":11,"groups":[{"kind":"contests","label":"Contests","papers":1,"sentences":1,"quotes":[{"arxiv_id":"1508.04395","title":"End-to-End Attention-based Large Vocabulary Speech Recognition","url":"/paper/1508.04395","section":"Related Work","sentence":"[ref] attempt to rectify this for speech recognition by using a moving windowed attention, but they do not provide a mechanism to address the situation that arises when no output can be produced from the windowed segment of data.","intent_p":0.81,"contests_p":0.79}]},{"kind":"method_or_tool","label":"Uses the method or tool of","papers":1,"sentences":1,"quotes":[{"arxiv_id":"1508.01211","title":"Listen, Attend and Spell","url":"/paper/1508.01211","section":"Computing $f_context$","sentence":"We first describe how the context vector is computed by an attention model similar to earlier work [ref] .","intent_p":0.93,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":6,"sentences":7,"quotes":[{"arxiv_id":"1506.06714","title":"A Neural Network Approach to Context-Sensitive Generation of Conversational Responses","url":"/paper/1506.06714","section":"Introduction","sentence":"The recently introduced sequence-to-sequence model has shown success in many tasks that map sequences to sequences, e.g., translation, speech recognition, image captioning and dialogue modeling [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"1508.01211","title":"Listen, Attend and Spell","url":"/paper/1508.01211","section":"Introduction","sentence":"The recently introduced sequence-to-sequence model has shown success in many tasks that map sequences to sequences, e.g., translation, speech recognition, image captioning and dialogue modeling [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"1511.04834","title":"Neural Programmer: Inducing Latent Programs with Gradient Descent","url":"/paper/1511.04834","section":"Related Work","sentence":"In the past few years, many proposals have been made to add more power or flexibility to neural networks, especially via the concept of augmented memory [ref] or augmented arithmetic units [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"1511.06279","title":"Neural Programmer-Interpreters","url":"/paper/1511.06279","section":"Related Work","sentence":"In the past few years, many proposals have been made to add more power or flexibility to neural networks, especially via the concept of augmented memory [ref] or augmented arithmetic units [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"1410.5401","title":"Neural Turing Machines","url":"/paper/1410.5401","section":"Related Work","sentence":"In the past few years, many proposals have been made to add more power or flexibility to neural networks, especially via the concept of augmented memory [ref] or augmented arithmetic units [ref] .","intent_p":1.0,"contests_p":null}]}],"unlabelled_sentences":2},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/1511.04868"}}