自然语言处理领域重要论文&资源全索引

标签: 自然语言 领域 论文 | 发表时间:2017-10-14 11:51 | 作者:刘晓坤
出处:https://www.jiqizhixin.com/
自然语言处理(NLP)是人工智能研究中极具挑战的一个分支。随着深度学习等技术的引入,NLP 领域正在以前所未有的速度向前发展。但对于初学者来说,这一领域目前有哪些研究和资源是必读的?最近,Kyubyong Park 为我们整理了一份完整列表。

 

GitHub 项目链接:https://github.com/Kyubyong/nlp_tasks

 

本人从事自然语言处理任务(NLP)的研究已经有很长时间了,有一天我想到,我需要为庞大的 NLP 领域做一个概览,我知道自己肯定不是想要一睹 NLP 任务的全貌的第一个人。

 

我曾竭尽所能的研究过尽可能多种类型的 NLP 任务,但由于个人知识的局限,我承认还远远没有穷尽整个领域。目前,该项目选取的参考文献都偏重最新的深度学习研究成果。我希望这些能为想要深入钻研一个 NLP 任务的人们提供一个开端。这个项目将持续更新,不过,我更希望与更多人合作。如果你有意愿的话,欢迎对这个项目作出贡献。

 

回指解析

 

  • See Coreference Resolution (https://github.com/Kyubyong/nlp_tasks#coreference-resolution)

 

自动作文评分

 

  • 论文:AutomaticText Scoring Using Neural Networks (https://arxiv.org/abs/1606.04289)
  • 论文:ANeural Approach to Automated Essay Scoring (http://www.aclweb.org/old_anthology/D/D16/D16-1193.pdf)
  • 竞赛:Kaggle:The Hewlett Foundation: Automated Essay Scoring (https://www.kaggle.com/c/asap-aes)
  • 项目:EnhancedAI Scoring Engine(https://github.com/edx/ease)

 

自动语音识别

 

  • WIKI Speech recognition(https://en.wikipedia.org/wiki/Speech_recognition)
  • 论文:DeepSpeech 2: End-to-End Speech Recognition in English and Mandarin (https://arxiv.org/abs/1512.02595)
  • 论文:WaveNet:A Generative Model for Raw Audio (https://arxiv.org/abs/1609.03499)
  • 项目:A TensorFlow implementation of Baidu's DeepSpeech architecture (https://github.com/mozilla/DeepSpeech)
  • 项目:Speech-to-Text-WaveNet: End-to-end sentence level English speech recognition using DeepMind's WaveNet(https://github.com/buriburisuri/speech-to-text-wavenet)
  • 竞赛:The 5thCHiME Speech Separation and Recognition Challenge (http://spandh.dcs.shef.ac.uk/chime_challenge/)
  • 资源:The 5thCHiME Speech Separation and Recognition Challenge (http://spandh.dcs.shef.ac.uk/chime_challenge/download.html)
  • 资源:CSTRVCTK Corpus (http://homepages.inf.ed.ac.uk/jyamagis/page3/page58/page58.html)
  • 资源:LibriSpeechASR corpus (http://www.openslr.org/12/)
  • 资源:Switchboard-1Telephone Speech Corpus (https://catalog.ldc.upenn.edu/ldc97s62)
  • 资源:TED-LIUMCorpus (http://www-lium.univ-lemans.fr/en/content/ted-lium-corpus)

 

自动摘要

 

  • WIKI Automatic summarization (https://en.wikipedia.org/wiki/Automatic_summarization)
  • 书籍:AutomaticText Summarization (https://www.amazon.com/Automatic-Text-Summarization-Juan-Manuel-Torres-Moreno/dp/1848216688/ref=sr_1_1?s=books&ie=UTF8&qid=1507782304&sr=1-1&keywords=Automatic+Text+Summarization)
  • 论文:TextSummarization Using Neural Networks (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.823.8025&rep=rep1&type=pdf)
  • 论文:Rankingwith Recursive Neural Networks and Its Application to Multi-DocumentSummarization (https://www.aaai.org/ocs/index.php/AAAI/AAAI15/paper/viewFile/9414/9520)
  • 资源:TextAnalytics Conferences(TAC)(https://tac.nist.gov/data/index.html)
  • 资源:DocumentUnderstanding Conferences (DUC)(http://www-nlpir.nist.gov/projects/duc/data.html)

 

指代消解

 

  • INFO Coreference Resolution(https://nlp.stanford.edu/projects/coref.shtml)
  • 论文:DeepReinforcement Learning for Mention-Ranking Coreference Models (https://arxiv.org/abs/1609.08667)
  • 论文:ImprovingCoreference Resolution by Learning Entity-Level Distributed Representations(https://arxiv.org/abs/1606.01323)
  • 竞赛:CoNLL2012 Shared Task: Modeling Multilingual Unrestricted Coreference in OntoNotes(http://conll.cemantix.org/2012/task-description.html)
  • 竞赛:CoNLL2011 Shared Task: Modeling Unrestricted Coreference in OntoNotes (http://conll.cemantix.org/2011/task-description.html)

 

实体链接

 

  • 见「命名实体消歧」部分

 

语法错误纠正

 

  • 论文:NeuralNetwork Translation Models for Grammatical Error Correction (https://arxiv.org/abs/1606.00189)
  • 竞赛:CoNLL-2013Shared Task: Grammatical Error Correction (http://www.comp.nus.edu.sg/~nlp/conll13st.html)
  • 竞赛:CoNLL-2014Shared Task: Grammatical Error Correction (http://www.comp.nus.edu.sg/~nlp/conll14st.html)
  • 资源:NUSNon-commercial research/trial corpus license (http://www.comp.nus.edu.sg/~nlp/conll14st/nucle_license.pdf)
  • 资源:Lang-8Learner Corpora(http://cl.naist.jp/nldata/lang-8/)
  • 资源:CornellMovie--Dialogs Corpus (http://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html)
  • 项目:DeepText Corrector(https://github.com/atpaino/deep-text-corrector)
  • 产品:deepgrammar(http://deepgrammar.com/)

 

字素音素转换

 

  • 论文:Grapheme-to-PhonemeModels for (Almost) Any Language (https://pdfs.semanticscholar.org/b9c8/fef9b6f16b92c6859f6106524fdb053e9577.pdf)
  • 论文:PolyglotNeural Language Models: A Case Study in Cross-Lingual Phonetic RepresentationLearning (https://arxiv.org/pdf/1605.03832.pdf)
  • 论文:MultitaskSequence-to-Sequence Models for Grapheme-to-Phoneme Conversion (https://pdfs.semanticscholar.org/26d0/09959fa2b2e18cddb5783493738a1c1ede2f.pdf)
  • 项目:Sequence-to-Sequence G2P toolkit (https://github.com/cmusphinx/g2p-seq2seq)
  • 资源:Multilingual Pronunciation Data (https://drive.google.com/drive/folders/0B7R_gATfZJ2aWkpSWHpXUklWUmM)

 

语种猜测

 

  • 见「语种辨别」部分

 

语种辨别

 

  • WIKI Language identification (https://en.wikipedia.org/wiki/Language_identification)
  • 论文:AUTOMATICLANGUAGE IDENTIFICATION USING DEEP NEURAL NETWORKS (https://repositorio.uam.es/bitstream/handle/10486/666848/automatic_lopez-moreno_ICASSP_2014_ps.pdf?sequence=1)
  • 竞赛: 2015Language Recognition Evaluation (https://www.nist.gov/itl/iad/mig/2015-language-recognition-evaluation)

 

语言建模

 

  • WIKI Language model (https://en.wikipedia.org/wiki/Language_model)
  • 工具包: KenLMLanguage Model Toolkit (http://kheafield.com/code/kenlm/)
  • 论文:DistributedRepresentations of Words and Phrases and their Compositionality (http://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf)
  • 论文:Character-AwareNeural Language Models (https://www.aaai.org/ocs/index.php/AAAI/AAAI16/paper/viewFile/12489/12017)
  • 资源: PennTreebank (https://github.com/townie/PTB-dataset-from-Tomas-Mikolov-s-webpage/tree/master/data)

 

语种识别

 

  • 见「语种辨别」部分

 

同一词类

 

  • WIKI Lemmatisation (https://en.wikipedia.org/wiki/Lemmatisation)
  • 论文: JointLemmatization and Morphological Tagging with LEMMING (http://www.cis.lmu.de/~muellets/pdf/emnlp_2015.pdf)
  • 工具包:WordNet Lemmatizer (http://www.nltk.org/api/nltk.stem.html#nltk.stem.wordnet.WordNetLemmatizer.lemmatize)
  • 资源:Treebank-3 (https://catalog.ldc.upenn.edu/ldc99t42)

 

观唇辨意

 

  • WIKI Lip reading (https://en.wikipedia.org/wiki/Lip_reading)
  • 论文:LipReading Sentences in the Wild (https://arxiv.org/abs/1611.05358)
  • 论文:3DConvolutional Neural Networks for Cross Audio-Visual Matching Recognition (https://arxiv.org/abs/1706.05739)
  • 项目: LipReading - Cross Audio-Visual Recognition using 3D Convolutional Neural Networks(https://github.com/astorfi/lip-reading-deeplearning)
  • 资源: TheGRID audiovisual sentence corpus (http://spandh.dcs.shef.ac.uk/gridcorpus/)

 

机器翻译

 

  • 论文:NeuralMachine Translation by Jointly Learning to Align and Translate (https://arxiv.org/abs/1409.0473)
  • 论文:NeuralMachine Translation in Linear Time (https://arxiv.org/abs/1610.10099)
  • 论文:AttentionIs All You Need (https://arxiv.org/abs/1706.03762)
  • 竞赛: ACL2014 NINTH WORKSHOP ON STATISTICAL MACHINE TRANSLATION (http://www.statmt.org/wmt14/translation-task.html#download)
  • 竞赛: EMNLP2017 SECOND CONFERENCE ON MACHINE TRANSLATION (WMT17)(http://www.statmt.org/wmt17/translation-task.html)
  • 资源:OpenSubtitles2016 (http://opus.lingfil.uu.se/OpenSubtitles2016.php)
  • 资源: WIT3:Web Inventory of Transcribed and Translated Talks (https://wit3.fbk.eu/)
  • 资源: TheQCRI Educational Domain (QED) Corpus (http://alt.qcri.org/resources/qedcorpus/)

 

生成词法变化

 

  • WIKI Inflection (https://en.wikipedia.org/wiki/Inflection)
  • 论文:MorphologicalInflection Generation Using Character Sequence to Sequence Learning (https://arxiv.org/abs/1512.06110)
  • 竞赛:SIGMORPHON 2016 Shared Task: Morphological Reinflection (http://ryancotterell.github.io/sigmorphon2016/)
  • 资源:sigmorphon2016 (https://github.com/ryancotterell/sigmorphon2016)

 

命名实体消歧

 

  • WIKI Entity linking (https://en.wikipedia.org/wiki/Entity_linking)
  • 论文:Robustand Collective Entity Disambiguation through Semantic Embeddings (http://www.stefanzwicklbauer.info/pdf/Sigir_2016.pdf)

 

命名实体识别

 

  • WIKI Named-entity recognition (https://en.wikipedia.org/wiki/Named-entity_recognition)
  • 论文:NeuralArchitectures for Named Entity Recognition (https://arxiv.org/abs/1603.01360)
  • 项目: OSUTwitter NLP Tools (https://github.com/aritter/twitter_nlp)
  • 竞赛: NamedEntity Recognition in Twitter (https://noisy-text.github.io/2016/ner-shared-task.html)
  • 竞赛: CoNLL2002 Language-Independent Named Entity Recognition (https://www.clips.uantwerpen.be/conll2002/ner/)
  • 竞赛:Introduction to the CoNLL-2003 Shared Task: Language-Independent Named EntityRecognition (http://aclweb.org/anthology/W03-0419)
  • 资源:CoNLL-2002 NER corpus (https://github.com/teropa/nlp/tree/master/resources/corpora/conll2002)
  • 资源:CoNLL-2003 NER corpus (https://github.com/synalp/NER/tree/master/corpus/CoNLL-2003)
  • 资源: NUTNamed Entity Recognition in Twitter Shared task (https://github.com/aritter/twitter_nlp/tree/master/data/annotated/wnut16)

 

释义检测

 

  • 论文:DynamicPooling and Unfolding Recursive Autoencoders for Paraphrase Detection (http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.650.7199&rep=rep1&type=pdf)
  • 项目:Paralex: Paraphrase-Driven Learning for Open Question Answering (http://knowitall.cs.washington.edu/paralex/)
  • 资源:Microsoft Research Paraphrase Corpus (https://www.microsoft.com/en-us/download/details.aspx?id=52398)
  • 资源:Microsoft Research Video Description Corpus (https://www.microsoft.com/en-us/download/details.aspx?id=52422&from=http%3A%2F%2Fresearch.microsoft.com%2Fen-us%2Fdownloads%2F38cf15fd-b8df-477e-a4e4-a4680caa75af%2F)
  • 资源: PascalDataset (http://nlp.cs.illinois.edu/HockenmaierGroup/pascal-sentences/index.html)
  • 资源:Flicker Dataset (http://nlp.cs.illinois.edu/HockenmaierGroup/8k-pictures.html)
  • 资源: TheSICK data set (http://clic.cimec.unitn.it/composes/sick.html)
  • 资源: PPDB:The Paraphrase Database (http://www.cis.upenn.edu/~ccb/ppdb/)
  • 资源:WikiAnswers Paraphrase Corpus (http://knowitall.cs.washington.edu/paralex/wikianswers-paraphrases-1.0.tar.gz)

 

语法分析

 

  • WIKI Parsing (https://en.wikipedia.org/wiki/Parsing)
  • 工具包: TheStanford Parser: A statistical parser (https://nlp.stanford.edu/software/lex-parser.shtml)
  • 工具包: spaCyparser (https://spacy.io/docs/usage/dependency-parse)
  • 论文:A fastand accurate dependency parser using neural networks (http://www.aclweb.org/anthology/D14-1082)
  • 竞赛: CoNLL2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies (http://universaldependencies.org/conll17/)
  • 竞赛: CoNLL2016 Shared Task: Multilingual Shallow Discourse Parsing (http://www.cs.brandeis.edu/~clp/conll16st/)
  • 竞赛: CoNLL2015 Shared Task: Shallow Discourse Parsing (http://www.cs.brandeis.edu/~clp/conll15st/)
  • 竞赛:SemEval-2016 Task 8: The meaning representations may be abstract, but this taskis concrete! (http://alt.qcri.org/semeval2016/task8/)

 

词性标记

 

  • WIKI Part-of-speech tagging (https://en.wikipedia.org/wiki/Part-of-speech_tagging)
  • 论文:MultilingualPart-of-Speech Tagging with Bidirectional Long Short-Term Memory Models andAuxiliary Loss (https://arxiv.org/pdf/1604.05529.pdf)
  • 论文:UnsupervisedPart-Of-Speech Tagging with Anchor Hidden Markov Models (https://transacl.org/ojs/index.php/tacl/article/viewFile/837/192)
  • 资源:Treebank-3 (https://catalog.ldc.upenn.edu/ldc99t42)
  • 工具包:nltk.tag package (http://www.nltk.org/api/nltk.tag.html)

 

拼音-中文转换

 

  • 论文:NeuralNetwork Language Model for Chinese Pinyin Input Method Engine (http://aclweb.org/anthology/Y15-1052)
  • 项目: NeuralChinese Transliterator (https://github.com/Kyubyong/neural_chinese_transliterator)

 

问答系统

 

  • WIKI Question answering (https://en.wikipedia.org/wiki/Question_answering)
  • 论文:Ask MeAnything: Dynamic Memory Networks for Natural Language Processing (http://www.thespermwhale.com/jaseweston/ram/papers/paper_21.pdf)
  • 论文:DynamicMemory Networks for Visual and Textual Question Answering (http://proceedings.mlr.press/v48/xiong16.pdf)
  • 竞赛: TRECQuestion Answering Task (http://trec.nist.gov/data/qamain.html)
  • 竞赛:NTCIR-8: Advanced Cross-lingual Information Access (ACLIA)(http://aclia.lti.cs.cmu.edu/ntcir8/Home)
  • 竞赛: CLEFQuestion Answering Track (http://nlp.uned.es/clef-qa/)
  • 竞赛:SemEval-2017 Task 3: Community Question Answering (http://alt.qcri.org/semeval2017/task3/)
  • 资源: MSMARCO: Microsoft MAchine Reading COmprehension Dataset (http://www.msmarco.org/)
  • 资源:Maluuba NewsQA (https://github.com/Maluuba/newsqa)
  • 资源: SQuAD:100,000+ Questions for Machine Comprehension of Text (https://rajpurkar.github.io/SQuAD-explorer/)
  • 资源:GraphQuestions: A Characteristic-rich Question Answering Dataset (https://github.com/ysu1989/GraphQuestions)
  • 资源: StoryCloze Test and ROCStories Corpora (http://cs.rochester.edu/nlp/rocstories/)
  • 资源:Microsoft Research WikiQA Corpus (https://www.microsoft.com/en-us/download/details.aspx?id=52419&from=http%3A%2F%2Fresearch.microsoft.com%2Fen-us%2Fdownloads%2F4495da01-db8c-4041-a7f6-7984a4f6a905%2Fdefault.aspx)
  • 资源:DeepMind Q&A Dataset (http://cs.nyu.edu/~kcho/DMQA/)
  • 资源: QASent(http://cs.stanford.edu/people/mengqiu/data/qg-emnlp07-data.tgz)

 

关系提取

 

  • WIKI Relationship extraction (https://en.wikipedia.org/wiki/Relationship_extraction)
  • 论文:A deeplearning approach for relationship extraction from interaction context insocial manufacturing paradigm (http://www.sciencedirect.com/science/article/pii/S0950705116001210)

 

语义角色标注

 

  • WIKI Semantic role labeling (https://en.wikipedia.org/wiki/Semantic_role_labeling)
  • 书籍:Semantic Role Labeling (https://www.amazon.com/Semantic-Labeling-Synthesis-Lectures-Technologies/dp/1598298313/ref=sr_1_1?s=books&ie=UTF8&qid=1507776173&sr=1-1&keywords=Semantic+Role+Labeling)
  • 论文:End-to-endLearning of Semantic Role Labeling Using Recurrent Neural Networks (http://www.aclweb.org/anthology/P/P15/P15-1109.pdf)
  • 论文:NeuralSemantic Role Labeling with Dependency Path Embeddi ngs (https://arxiv.org/abs/1605.07515)
  • 竞赛:CoNLL-2005 Shared Task: Semantic Role Labeling (http://www.cs.upc.edu/~srlconll/st05/st05.html)
  • 竞赛:CoNLL-2004 Shared Task: Semantic Role Labeling (http://www.cs.upc.edu/~srlconll/st04/st04.html)
  • 工具包:Illinois Semantic Role Labeler (SRL)(http://cogcomp.org/page/software_view/SRL)
  • 资源:CoNLL-2005 Shared Task: Semantic Role Labeling (http://www.cs.upc.edu/~srlconll/soft.html)

 

语句边界消歧

 

  • WIKI Sentence boundary disambiguation (https://en.wikipedia.org/wiki/Sentence_boundary_disambiguation)
  • 论文:AQuantitative and Qualitative Evaluation of Sentence Boundary Detection for theClinical Domain (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5001746/)
  • 工具包: NLTKTokenizers (http://www.nltk.org/_modules/nltk/tokenize.html)
  • 资源: TheBritish National Corpus (http://www.natcorp.ox.ac.uk/)
  • 资源:Switchboard-1 Telephone Speech Corpus (https://catalog.ldc.upenn.edu/ldc97s62)

 

情绪分析

 

  • WIKI Sentiment analysis (https://en.wikipedia.org/wiki/Sentiment_analysis)
  • INFO Awesome Sentiment Analysis (https://github.com/xiamx/awesome-sentiment-analysis)
  • 竞赛:Kaggle: UMICH SI650 - Sentiment Classification (https://www.kaggle.com/c/si650winter11#description)
  • 竞赛:SemEval-2017 Task 4: Sentiment Analysis in Twitter (http://alt.qcri.org/semeval2017/task4/)
  • 竞赛:SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Microblogsand News (http://alt.qcri.org/semeval2017/task5/)
  • 项目:SenticNet (http://sentic.net/about/)
  • 资源:Multi-Domain Sentiment Dataset (version2.0)(http://www.cs.jhu.edu/~mdredze/datasets/sentiment/)
  • 资源:Stanford Sentiment Treebank (https://nlp.stanford.edu/sentiment/code.html)
  • 资源:Twitter Sentiment Corpus (http://www.sananalytics.com/lab/twitter-sentiment/)
  • 资源:Twitter Sentiment Analysis Training Corpus (http://thinknook.com/twitter-sentiment-analysis-training-corpus-dataset-2012-09-22/)
  • 资源: AFINN:List of English words rated for valence (http://www2.imm.dtu.dk/pubdb/views/publication_details.php?id=6010)

 

源分离

 

  • WIKI Source separation (https://en.wikipedia.org/wiki/Source_separation)
  • 论文:FromBlind to Guided Audio Source Separation (https://hal-univ-rennes1.archives-ouvertes.fr/hal-00922378/document)
  • 论文:JointOptimization of Masks and Deep Recurrent Neural Networks for Monaural SourceSeparation (https://arxiv.org/abs/1502.04149)
  • 竞赛: SignalSeparation Evaluation Campaign (SiSEC)(https://sisec.inria.fr/)
  • 竞赛: CHiMESpeech Separation and Recognition Challenge (http://spandh.dcs.shef.ac.uk/chime_challenge/)

 

说话人认证

 

  • 见「说话人识别」部分

 

语音身份分离

 

  • WIKI Speaker diarisation (https://en.wikipedia.org/wiki/Speaker_diarisation)
  • 论文:DNN-basedspeaker clustering for speaker diarisation (http://eprints.whiterose.ac.uk/109281/1/milner_is16.pdf)
  • 论文:UnsupervisedMethods for Speaker Diarization: An Integrated and Iterative Approach (http://groups.csail.mit.edu/sls/publications/2013/Shum_IEEE_Oct-2013.pdf)
  • 论文:Audio-VisualSpeaker Diarization Based on Spatiotemporal Bayesian Fusion (https://arxiv.org/pdf/1603.09725.pdf)
  • 竞赛: RichTranscription Evaluation (https://www.nist.gov/itl/iad/mig/rich-transcription-evaluation)

 

说话人识别

 

  • WIKI Speaker recognition (https://en.wikipedia.org/wiki/Speaker_recognition)
  • 论文:A NOVELSCHEME FOR SPEAKER RECOGNITION USING A PHONETICALLY-AWARE DEEP NEURAL NETWORK (https://pdfs.semanticscholar.org/204a/ff8e21791c0a4113a3f75d0e6424a003c321.pdf)
  • 论文:DEEPNEURAL NETWORKS FOR SMALL FOOTPRINT TEXT-DEPENDENT SPEAKER VERIFICATION (https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/41939.pdf)
  • 竞赛: NISTSpeaker Recognition Evaluation (SRE)(https://www.nist.gov/itl/iad/mig/speaker-recognition)
  • INFO Are there any suggestions for free databases for speakerrecognition? (https://www.researchgate.net/post/Are_there_any_suggestions_for_free_databases_for_speaker_recognition)

 

唇读

 

  • 见「观唇辨意」部分

 

语音识别

 

  • 见「自动语音识别」部分

 

语音分割

 

  • WIKI Speech_segmentation (https://en.wikipedia.org/wiki/Speech_segmentation)
  • 论文:WordSegmentation by 8-Month-Olds: When Speech Cues Count More Than Statistics (http://www.utm.toronto.edu/infant-child-centre/sites/files/infant-child-centre/public/shared/elizabeth-johnson/Johnson_Jusczyk.pdf)
  • 论文:UnsupervisedWord Segmentation and Lexicon Discovery Using Acoustic Word Embeddings (https://arxiv.org/abs/1603.02845)
  • 论文:UnsupervisedLexicon Discovery from Acoustic Inpu (http://www.aclweb.org/old_anthology/Q/Q15/Q15-1028.pdf)
  • 论文:Weaklysupervised spoken term discovery using cross-lingual side information (http://www.research.ed.ac.uk/portal/files/29957958/1609.06530v1.pdf)
  • 资源:CALLHOME Spanish Speech (https://catalog.ldc.upenn.edu/ldc96s35)

 

语音合成

 

  • WIKI Speech synthesis (https://en.wikipedia.org/wiki/Speech_synthesis)
  • 论文:WaveNet:A Generative Model for Raw Audio (https://arxiv.org/abs/1609.03499)
  • 论文:Tacotron:Towards End-to-End Speech Synthesis (https://arxiv.org/abs/1703.10135)
  • 论文:DeepVoice 2: Multi-Speaker Neural Text-to-Speech (https://arxiv.org/abs/1705.08947)
  • 资源: TheWorld English Bible (https://github.com/Kyubyong/tacotron)
  • 资源: LJSpeech Dataset (https://github.com/keithito/tacotron)
  • 资源: LessacData (http://www.cstr.ed.ac.uk/projects/blizzard/2011/lessac_blizzard2011/)
  • 竞赛:Blizzard Challenge 2017 (https://synsig.org/index.php/Blizzard_Challenge_2017)
  • PRODUCT Lyrebird (https://lyrebird.ai/)
  • 项目: TheFestvox project (http://www.festvox.org/index.html)
  • 工具包:Merlin: The Neural Network (NN) based Speech Synthesis System (https://github.com/CSTR-Edinburgh/merlin)

 

语音增强

 

  • WIKI Speech enhancement (https://en.wikipedia.org/wiki/Speech_enhancement)
  • 书籍: Speechenhancement: theory and practice (https://www.amazon.com/Speech-Enhancement-Theory-Practice-Second/dp/1466504218/ref=sr_1_1?ie=UTF8&qid=1507874199&sr=8-1&keywords=Speech+enhancement%3A+theory+and+practice)
  • 论文 AnExperimental Study on Speech Enhancement BasedonDeepNeuralNetwork (http://staff.ustc.edu.cn/~jundu/Speech%20signal%20processing/publications/SPL2014_Xu.pdf)
  • 论文: ARegression Approach to Speech Enhancement BasedonDeepNeuralNetworks (https://www.researchgate.net/profile/Yong_Xu63/publication/272436458_A_Regression_Approach_to_Speech_Enhancement_Based_on_Deep_Neural_Networks/links/57fdfdda08aeaf819a5bdd97.pdf)
  • 论文: SpeechEnhancement Based on Deep Denoising Autoencoder (https://www.researchgate.net/profile/Yu_Tsao/publication/283600839_Speech_enhancement_based_on_deep_denoising_Auto-Encoder/links/577b486108ae213761c9c7f8/Speech-enhancement-based-on-deep-denoising-Auto-Encoder.pdf)

 

语音文本转换

 

  • 见「自动语音识别」部分

 

口语的术语检测

 

  • 见「语音分割」部分

 

词干提取

 

  • WIKI Stemming (https://en.wikipedia.org/wiki/Stemming)
  • 论文: ABACKPROPAGATION NEURAL NETWORK TO IMPROVE ARABIC STEMMING (http://www.jatit.org/volumes/Vol82No3/7Vol82No3.pdf)
  • 工具包: NLTKStemmers (http://www.nltk.org/howto/stem.html)

 

术语提取

 

  • WIKI Terminology extraction (https://en.wikipedia.org/wiki/Terminology_extraction)
  • 论文: NeuralAttention Models for Sequence Classification: Analysis and Application to KeyTerm Extraction and Dialogue Act Detection (https://arxiv.org/pdf/1604.00077.pdf)

 

文本简化

 

  • WIKI Text simplification (https://en.wikipedia.org/wiki/Text_simplification)
  • 论文:Aligning Sentences from Standard Wikipedia to Simple Wikipedia (https://ssli.ee.washington.edu/~hannaneh/papers/simplification.pdf)
  • 论文:Problems in Current Text Simplification Research: New Data Can Help (https://pdfs.semanticscholar.org/2b8d/a013966c0c5e020ebc842d49d8ed166c8783.pdf)
  • 资源:Newsela Data (https://newsela.com/data/)

 

文本语音转换

 

  • 见「语音合成」部分

 

文本蕴涵

 

  • WIKI Textual entailment (https://en.wikipedia.org/wiki/Textual_entailment)
  • 项目:Textual Entailment with TensorFlow (https://github.com/Steven-Hewitt/Entailment-with-Tensorflow)
  • 论文:Textual Entailment with Structured Attentions and Composition (https://arxiv.org/pdf/1701.01126.pdf)
  • 竞赛:SemEval-2014 Task 1: Evaluation of compositional distributional semantic modelson full sentences through semantic relatedness and textual entailment (http://alt.qcri.org/semeval2014/task1/)
  • 竞赛:SemEval-2013 Task 7: The Joint Student Response Analysis and 8th RecognizingTextual Entailment Challenge (https://www.cs.york.ac.uk/semeval-2013/task7.html)

 

声音转换

 

  • 论文:PHONETIC POSTERIORGRAMS FOR MANY-TO-ONE VOICE CONVERSION WITHOUT PARALLEL DATATRAINING (http://www1.se.cuhk.edu.hk/~hccl/publications/pub/2016_paper_297.pdf)
  • 项目: Animplementation of voice conversion system utilizing phonetic posteriorgrams (https://github.com/sesenosannko/ppg_vc)
  • 竞赛: VoiceConversion Challenge 2016 (http://www.vc-challenge.org/vcc2016/index.html)
  • 竞赛: VoiceConversion Challenge 2018 (http://www.vc-challenge.org/)
  • 资源:CMU_ARCTIC speech synthesis databases (http://festvox.org/cmu_arctic/)
  • 资源: TIMITAcoustic-Phonetic Continuous Speech Corpus (https://catalog.ldc.upenn.edu/ldc93s1)

 

声音识别

 

  • 见「说话人识别」部分

 

词嵌入

 

  • WIKI Word embedding (https://en.wikipedia.org/wiki/Word_embedding)
  • 工具包:Gensim: word2vec (https://radimrehurek.com/gensim/models/word2vec.html)
  • 工具包:fastText (https://github.com/facebookresearch/fastText)
  • 工具包: GloVe:Global Vectors for Word Representation (https://nlp.stanford.edu/projects/glove/)
  • INFO Where to get a pretrained model (https://github.com/3Top/word2vec-api)
  • 项目:Pre-trained word vectors of 30+ languages (https://github.com/Kyubyong/wordvectors)
  • 项目:Polyglot: Distributed word representations for multilingual NLP (https://sites.google.com/site/rmyeid/projects/polyglot)

 

词预测

 

  • INFO What is Word Prediction? (http://www2.edc.org/ncip/library/wp/what_is.htm)
  • 论文: Theprediction of character based on recurrent neural network language model (http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7960065)
  • 论文: AnEmbedded Deep Learning based Word Prediction (https://arxiv.org/abs/1707.01662)
  • 论文:Evaluating Word Prediction: Framing Keystroke Savings (http://aclweb.org/anthology/P08-2066)
  • 资源: AnEmbedded Deep Learning based Word Prediction (https://github.com/Meinwerk/WordPrediction/master.zip)
  • 项目: WordPrediction using Convolutional Neural Networks—can you do better than iPhone™Keyboard? (https://github.com/Kyubyong/word_prediction)

 

词分割

 

  • WIKI Word segmentation (https://en.wikipedia.org/wiki/Text_segmentation#Segmentation_problems)
  • 论文: NeuralWord Segmentation Learning for Chinese (https://arxiv.org/abs/1606.04300)
  • 项目:Convolutional neural network for Chinese word segmentation (https://github.com/chqiwang/convseg)
  • 工具包:Stanford Word Segmenter (https://nlp.stanford.edu/software/segmenter.html)
  • 工具包: NLTKTokenizers (http://www.nltk.org/_modules/nltk/tokenize.html)

 

词义消歧

 

  • 资源:Word-sense disambiguation (https://en.wikipedia.org/wiki/Word-sense_disambiguation)
  • 论文:Train-O-Matic: Large-Scale Supervised Word Sense Disambiguation in MultipleLanguages without Manual Training Data (http://www.aclweb.org/anthology/D17-1008)
  • 资源:Train-O-Matic Data (http://trainomatic.org/data/train-o-matic-data.zip)
  • 资源:BabelNet (http://babelnet.org/)

相关 [自然语言 领域 论文] 推荐:

自然语言处理领域重要论文&资源全索引

- - 机器之心
自然语言处理(NLP)是人工智能研究中极具挑战的一个分支. 随着深度学习等技术的引入,NLP 领域正在以前所未有的速度向前发展. 但对于初学者来说,这一领域目前有哪些研究和资源是必读的. 最近,Kyubyong Park 为我们整理了一份完整列表. GitHub 项目链接:https://github.com/Kyubyong/nlp_tasks.

自然语言处理概览

- - 互联网 - ITeye博客
自然语言处理研究的是词、句、文档等几个层次的内容. 一、发展历史(前两个是理论、后一个是经验). 1、形式语法(复杂特征集). 2、词汇主义方法(WordNet、ConceptNet、FrameNet), 人工总结和整理概念、层次、结构等 . 3、统计语言模型(语言有统计规律性,让机器去自己学习规律).

自然语言处理某个pipeline

- - 互联网 - ITeye博客
1、数据源:包括文本、pdf、数据库等不同来源. 2、使用到的库:jieba gensim sklearn keras . 3、可以实现的服务:找出相关和相近词(以分词为准)、比较2个分词的相似度、和哪些相关同时和别的不相关(语义上的模糊查找). 数据源的load->gensim->classifier(传统基于词频的/深度学习的 keras).

分布式系统领域经典论文翻译集

- khsing - 弯曲评论
优秀的不仅仅是能把这些经典论文看完,而是还花心思翻译出来. 其联系方式是phylips@bmy】. 1.      google系列论文译序. 3.      面向星球的网络搜索:google集群架构. 4.      GFS:google文件系统. google系列论文翻译集(合集). 1.      分布式理论系列译序.

AI领域必看的45篇论文(附下载地址)

- - 机器之心
而AI领域的发展会是IT中最快的. 我们所看到的那些黑客技,其后面无不堆积了大量的论文. 从某种调度来讲,他们所用的技术跟书籍里的内容确实不是一个时代. 要想与时俱进,就必须改变思路——从论文入手. 今天给大家介绍45篇让你跟上AI时代的论文. No1  wide_deep模型论文:. 关于神经元、全连接网络之类的基础结构,想必每个AI学者都有了解.

统计自然语言处理的基础学习之一

- - CSDN博客互联网推荐文章
理性主义:其实就是纯粹使用规则的方法处理自然语言,并认为这些语言规则天生就存在人的基因中. 在计算机中重现这些规则,就能学会人的语言处理能力. 经验主义:认为人有感知和学习能力,通过概括、模式识别、联想等能力,来学习到自然语言的结构. 哲学上的问题,类似于起源之类,就先别考虑的. 统计方法在NLP中的地位是什么.

[微信机器人_04]自然语言处理简单实现

- - CSDN博客互联网推荐文章
这篇博文中将介绍奇迹蛋自然语言处理模块的实现,自然语言处理的三个关键词:分词、建库、匹配. 机器人要如何来记忆这两组对话呢?. 这样存储当然没有问题,但是显然太土了,有没有其它的办法呢. 在存储过程中,总是希望能够去除冗余,即相同的词语能够只存储一次. 很容易想到,把句子拆分成词来存储. 如图所示,每个问题都拆分成词,并以词为节点建立树.

Quora在自然语言处理上的所做的工作

- - 标点符
问答网站最核心的内容是提供给优质的内容,如何让内容更加优质,处理分析大量的文本数据也是必不可少的工作. Quora有大量的文本数据,分布在Quora的数百万个问题、答案和评论中. 不仅如此,还有大量的元数据来补充我问答,包括”赞”和”踩”,用户对话题的兴趣或擅长与否,问题和话题的关系,话题去重合并,用户的社交和关系和影响力幅射,以及用户在Quora的每一个操作历史.

自然语言处理词向量化总结

- - 冰火岛
distributed representation 分布式表达(一类表示方法,基于统计含义),分散式表达(从一个高维空间X映射到一个低维空间Y) 分布假说(distributional hypothesis)为这一设想提供了 理论基础:上下文相似的词,其语义也相似.. 自然语言处理的基础是词向量化,即文本数值化,后面进行数据挖掘工作就和常见的任务类似,即分类,聚类等等.