Spoken Language Understanding: Systems for Extracting Semantic Information from Speech

By Gökhan Tür, R. Mori. 2011

685 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/3863c181a54465b2fbd7af7fdb6fb165309f5296

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Jianfeng Gao
2. Yoshua Bengio
3. Li Deng
4. Xiaodong He
5. Mohit Bansal
6. Zhe Gan
7. Kyunghyun Cho
8. Bing Liu

Citations in the corpus

• Use of kernel deep convex networks and end-to-end learning for spoken language understanding (2012, background): “In both areas of intent determination (or domain detection) and slot filling, the recent state of art has been based on discriminative classifiers such as Boosting or SVM (Tur and De Mori, 2011) and CRF (Raymond and Riccardi 2007; Hahn et al, 2011).”
• Few-shot Slot Tagging with Collapsed Dependency Transfer and Label-enhanced Task-adaptive Projection Network (2020, background): “Slot tagging (Tur and De Mori, 2011), a key module in the task-oriented dialogue system (Young et al., 2013), is usually formulated as a sequence labeling problem (Sarikaya et al., 2016).”
• Using Recurrent Neural Networks for Slot Filling in Spoken Language Understanding (2015, unknown): “at machines [1].”
• The Technology Behind Personal Digital Assistants: An overview of the system architecture and key components (2017, background): “The LU component takes the speech transcription (or the text input if the user types) and performs a semantic analysis to determine the underlying user intent [11]–[14], [17].”
• A Stack-Propagation Framework with Token-Level Intent Detection for Spoken Language Understanding (2019, background): “It usually consists of intent detection to identify users’ intents and slot filling task to extract semantic constituents from the natural language utterances (Tur and De Mori, 2011).”
• Survey on evaluation methods for dialogue systems (2019, background): “Since the early 2000s, the natural language understanding task is often seen as a set of subtasks (Tur and Mori 2011) as follows: (i) identification of domain (if multiple domains), (ii) identification of intents (that is, the question type, the dialogue act, etc.) and (iii) identification of the…”
• CG-BERT: Conditional Text Generation with BERT for Generalized Few-shot Intent Detection (2020, background): “Intent Detection (ID) is a key task in spoken dialogue systems, which aims at understanding the in-tents behind users’ inputs (Tur and De Mori, 2011).”
• Multi-Domain Joint Semantic Frame Parsing Using Bi-Directional RNN-LSTM (2016, background): “Three key tasks in such targeted understanding applications are domain classification, intent determination and slot filling [1], aiming to form a semantic frame that captures the semantics of user utterances/queries.”
• Investigation of recurrent-neural-network architectures and learning methods for spoken language understanding (2013, background): “Despite many years of research, the slot filling task in SLU is still a challenging problem, in parallel with the intent determination task [23][27].”
• SPLAT: Speech-Language Joint Pre-Training for Spoken Language Understanding (2021, background): “SLU has been widely employed in various areas such as intent understanding (Tur and De Mori, 2011; Bhargava et al., 2013; Ravuri and Stolcke, 2015; Lugosch et al., 2019), question answering (Lee et al., 2018; Chuang et al., 2020), and sentiment analysis (Zadeh et al., 2018).”
• Efficient Large-Scale Domain Classification with Personalized Attention (2018, background): “Traditional multi-domain SLU/NLU systems are designed hierarchically, starting with domain clas-sification to classify an incoming utterance into one of many possible domains, followed by further semantic analysis with domain-specific intent classification and slot tagging (Tur and de Mori, 2011).”
• Detecting out-of-domain utterances addressed to a virtual personal assistant (2014, background): “Spoken language understanding (SLU) in human/machine spoken dialog systems aims to automatically identify the domain and intent of the user as expressed in natural language (NL) and to extract associated arguments or slots [1] to achieve a goal.”
• Application of Deep Belief Networks for Natural Language Understanding (2014, unknown): “Boosting has been applied to a number of natural language processing tasks in the past [9].”
• Speech Model Pre-training for End-to-End Spoken Language Understanding (2019, background): “Spoken language understanding (SLU) systems infer the meaning or intent of a spoken utterance [1].”
• From Eliza to XiaoIce: challenges and opportunities with social chatbots (2018, background): “These systems usually operate on constrained domains (Glass et al., 1995; Walk et al., 2001; Raux et al., 2005; Andreani et al., 2006; Wang et al., 2011; Tur and Mori, 2011).”