An Interactive Dialogue Management System for Spoken Language

The spoken dialogue system is one of the most important human-machine communication ways. Human-machine communication can be described as an interaction between the user and the computer. This field is full of research points, so it is considered a good attractive environment for many researchers. The spoken dialogue system is of great importance in the process of communicating commercial applications, and facilitating the connecting process between the human and machine which may take different faces. The main objective of this research will be building an interactive dialogue management system for spoken dialogue system in an ideal way, By answering the following main question: How can we build an interactive dialogue management system for spoken dialogue system in an ideal way has the ability to accomplish the Naturalness, Usability, Mixed initiative, Co-operativity, Robustness, and Exploration. This research will be a mixed-method research and will adopt a descriptive survey design in collecting information by Survey questionnaires, Interviews to a sample of the target population, and while secondary data will be found from books, journals, and The Internet. The most important conclusion of the research is the spoken dialogue system is to be less complexity and use uncertainty model; this way must be acceptable by the user and the system itself.

Two-stage Behavior Cloning for Spoken Dialogue System in Debt Collection

With the rapid growth of internet finance and the booming of financial lending, the intelligent calling for debt collection in FinTech companies has driven increasing attention. Nowadays, the widely used intelligent calling system is based on dialogue flow, namely configuring the interaction flow with the finite-state machine. In our scenario of debt collection, the completed dialogue flow contains more than one thousand interactive paths. All the dialogue procedures are artificially specified, with extremely high maintenance costs and error-prone. To solve this problem, we propose the behavior-cloning-based collection robot framework without any dialogue flow configuration, called two-stage behavior cloning (TSBC). In the first stage, we use multi-label classification model to obtain policies that may be able to cope with the current situation according to the dialogue state; in the second stage, we score several scripts under each obtained policy to select the script with the highest score as the reply for the current state. This framework makes full use of the massive manual collection records without labeling and fully absorbs artificial wisdom and experience. We have conducted extensive experiments in both single-round and multi-round scenarios and showed the effectiveness of the proposed system. The accuracy of a single round of dialogue can be improved by 5%, and the accuracy of multiple rounds of dialogue can be increased by 3.1%.

Using Complexity-Identical Human- and Machine-Directed Utterances to Investigate Addressee Detection for Spoken Dialogue Systems

Human-machine addressee detection (H-M AD) is a modern paralinguistics and dialogue challenge that arises in multiparty conversations between several people and a spoken dialogue system (SDS) since the users may also talk to each other and even to themselves while interacting with the system. The SDS is supposed to determine whether it is being addressed or not. All existing studies on acoustic H-M AD were conducted on corpora designed in such a way that a human addressee and a machine played different dialogue roles. This peculiarity influences speakers’ behaviour and increases vocal differences between human- and machine-directed utterances. In the present study, we consider the Restaurant Booking Corpus (RBC) that consists of complexity-identical human- and machine-directed phone calls and allows us to eliminate most of the factors influencing speakers’ behaviour implicitly. The only remaining factor is the speakers’ explicit awareness of their interlocutor (technical system or human being). Although complexity-identical H-M AD is essentially more challenging than the classical one, we managed to achieve significant improvements using data augmentation (unweighted average recall (UAR) = 0.628) over native listeners (UAR = 0.596) and a baseline classifier presented by the RBC developers (UAR = 0.539).

An Integrated Process Based Natural Language Processing System

Natural Language Processing (NLP) systems involve Natural Language Understanding (NLU), Dialogue Management (DM) and Natural Language Generation (NLG). The purpose of this work involves integrating learning with examples and rule-based processing to design an NLP system. The design involves a three-stage processing framework, which combines syntactic generation, semantic extraction and a strong rule-based control. The syntactic generator generates syntax by aligning sentences with Part-of-Speech (POS) tags limited by the number of words in the lexicon. The semantic extractor extracts meaningful keywords from the queries raised. The above two modules are controlled by generalized rules by the rule-based controller module. The system is evaluated under different domains. The results reveal that the accuracy of the system is 92.33% on an average. The design process is simple, and the processing time is 2.12 seconds, which is minimal compared to similar statistical models. The performance of an NLP tool in a certain task can be estimated by the quality of its predictions on the classification of unseen data. The results reveal similar performance with existing systems indicating the possibility of usage for similar tasks. The system supports a vocabulary of about 700 words and can be used as an NLP module in a spoken dialogue system for various domains or task areas.

Isolated Word based Spoken Dialogue System using Odia Phones

: In this research paper, we have developed a spoken dialogue system using Odia phone set. We have also added additional security feature to our developed spoken dialogue system by integrating with speaker verification module, which allows the services to only genuine users. The spoken dialogue system can give the bouquet of services relating to opening of frequently usage applications, files and folders that are either installed or stored in user’s computers. The spoken dialogue system also responds to the users in synthesized speech mode relating to the service. The spoken dialogue system can be used to keep the desktop of computer from free of clutter. We have used HMM based Odia isolated word speech recognition engine and fuzzy c-means based speaker verification module in development of spoken dialogue system. The accuracy of Odia speech recognition engine is found as 78.22 % and 62.31% for seen and unseen users respectively and the average accuracy rate of speaker verification module is found as 66.2%.