| { |
| "paper_id": "O18-1013", |
| "header": { |
| "generated_with": "S2ORC 1.0.0", |
| "date_generated": "2023-01-19T08:09:46.455446Z" |
| }, |
| "title": "Weather Forecast Voice System", |
| "authors": [ |
| { |
| "first": "Thanh", |
| "middle": [], |
| "last": "Do", |
| "suffix": "", |
| "affiliation": { |
| "laboratory": "Languages and Information Technology (HUFLIT)", |
| "institution": "Ho Chi Minh City University of Foreign", |
| "location": {} |
| }, |
| "email": "dodinhthanh@huflit.edu.vn" |
| } |
| ], |
| "year": "", |
| "venue": null, |
| "identifiers": {}, |
| "abstract": "This paper introduces Weather Voice System applied to information commands-responses system through Public Switched Telephone Network (PSTN). Being able to understand users' voice commands, our system would assist them during their searching weather forecast in Vietnamese towns and cities by communicating in Vietnamese through Telephone Network. What makes this system different from others consists of its ability to analyze syntax and semantics of voice commands after their being recognized by speech recognizer component. Weather Voice is composed of main components that are there: 1) Telephone network communicator, 2) Vietnamese speech recognizer, 3) Natural language processor, and 4) Vietnamese speech synthesizer. To the best of our knowledge, this is one of the first systems in Vietnam to combine text language processing and speech processing. This event could make voice applications more intelligent with aim to communicate with humans in natural language at high precision and at quick speed. The fact that experiments have proved our expectation as well as the friendliness of the system to users reveals the practicality of our research.", |
| "pdf_parse": { |
| "paper_id": "O18-1013", |
| "_pdf_hash": "", |
| "abstract": [ |
| { |
| "text": "This paper introduces Weather Voice System applied to information commands-responses system through Public Switched Telephone Network (PSTN). Being able to understand users' voice commands, our system would assist them during their searching weather forecast in Vietnamese towns and cities by communicating in Vietnamese through Telephone Network. What makes this system different from others consists of its ability to analyze syntax and semantics of voice commands after their being recognized by speech recognizer component. Weather Voice is composed of main components that are there: 1) Telephone network communicator, 2) Vietnamese speech recognizer, 3) Natural language processor, and 4) Vietnamese speech synthesizer. To the best of our knowledge, this is one of the first systems in Vietnam to combine text language processing and speech processing. This event could make voice applications more intelligent with aim to communicate with humans in natural language at high precision and at quick speed. The fact that experiments have proved our expectation as well as the friendliness of the system to users reveals the practicality of our research.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Abstract", |
| "sec_num": null |
| } |
| ], |
| "body_text": [ |
| { |
| "text": "In the recent years, Vietnamese voice processing researches have found themselves making significant progress. Among them, two which must be mentioned are Voice recognizing ( conducted and published many works high estimated [2] , [4] , [6] , [8] , [14] . Especially in 2012, the authors from Natural Science University succeeded in building Voice Server System [9] to explore potential applications of Voice Processing whose use was made of by media systems, such as Voice Information Providing System through mobile and fixed phones, this work was awarded the second prize of Vietnamese Gifted. However, the works above only focused on improving the processing of Vietnamese voice and still neglected the semantic analyzing of voice commands.", |
| "cite_spans": [ |
| { |
| "start": 173, |
| "end": 174, |
| "text": "(", |
| "ref_id": null |
| }, |
| { |
| "start": 225, |
| "end": 228, |
| "text": "[2]", |
| "ref_id": "BIBREF1" |
| }, |
| { |
| "start": 231, |
| "end": 234, |
| "text": "[4]", |
| "ref_id": "BIBREF3" |
| }, |
| { |
| "start": 237, |
| "end": 240, |
| "text": "[6]", |
| "ref_id": "BIBREF6" |
| }, |
| { |
| "start": 243, |
| "end": 246, |
| "text": "[8]", |
| "ref_id": "BIBREF8" |
| }, |
| { |
| "start": 249, |
| "end": 253, |
| "text": "[14]", |
| "ref_id": "BIBREF14" |
| }, |
| { |
| "start": 362, |
| "end": 365, |
| "text": "[9]", |
| "ref_id": "BIBREF9" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1." |
| }, |
| { |
| "text": "Weather Voice is Voice Server System built on the combination of speech language processing and text language one. It can recognize numerous patterns of Vietnamese voice commands in order to transform them into text, then analyze their syntax and semantics, afterwards carry out searching data base, and eventually answer to users with the data searched in Vietnamese voice.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1." |
| }, |
| { |
| "text": "The analyzing syntax and semantics of commands is assisted by DCG (Definite Clause Grammar) [3] , [7] , [11] , [12] , [17] . During the voice processing, we call for HTK (Hidden Markov Model Toolkit) [13] to recognize voices and apply Unit -selection [1] method to synthesizing voices.", |
| "cite_spans": [ |
| { |
| "start": 92, |
| "end": 95, |
| "text": "[3]", |
| "ref_id": "BIBREF2" |
| }, |
| { |
| "start": 98, |
| "end": 101, |
| "text": "[7]", |
| "ref_id": "BIBREF7" |
| }, |
| { |
| "start": 104, |
| "end": 108, |
| "text": "[11]", |
| "ref_id": "BIBREF11" |
| }, |
| { |
| "start": 111, |
| "end": 115, |
| "text": "[12]", |
| "ref_id": "BIBREF12" |
| }, |
| { |
| "start": 118, |
| "end": 122, |
| "text": "[17]", |
| "ref_id": "BIBREF17" |
| }, |
| { |
| "start": 200, |
| "end": 204, |
| "text": "[13]", |
| "ref_id": "BIBREF13" |
| }, |
| { |
| "start": 251, |
| "end": 254, |
| "text": "[1]", |
| "ref_id": "BIBREF0" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1." |
| }, |
| { |
| "text": "The system is designed to achieve the following functions: Recognizing query statements through phones; analyzing them; searching data from database; answering to users through phones. Here is the scenario in details: (0) Standby state (1) User calls on the system and asks a command in Vietnamese.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "System structure", |
| "sec_num": "2." |
| }, |
| { |
| "text": "(2) The system reasserts the content of the user's command. _ (2.0) If the user confirms it, the system continues to carry out the step (3) _ (2.1) If the user disapproves of it, the system returns to the step (0) (3) The speech is transferred to the processing unit ASR and transformed into Vietnamese text.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "System structure", |
| "sec_num": "2." |
| }, |
| { |
| "text": "(4) The system analyzes the command's syntax and processes its semantic meaning. _ (4.0) If the command is grammatically correct * The system conducts searching database and answers to the user in speech. * When finishing, it returns to the step (0) _ (4.1) If it is not the case, which means that the command is grammatically incorrect, the system will ask the user to redo a command in speech.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "System structure", |
| "sec_num": "2." |
| }, |
| { |
| "text": "We will follow the example of a specific transaction: ", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "System structure", |
| "sec_num": "2." |
| }, |
| { |
| "text": "\u2212", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "System structure", |
| "sec_num": "2." |
| }, |
| { |
| "text": "In the Weather Voice system, we use HTK to build the voice recognizer component. HTK provides us with instruments analyzing voice, specially the one recognizing voice, based on HMM [13] . According to approaches of [9] , [10] , [15] , [16] , [18] we implement a contextdependent model making use of triphone to recognize words from word list as well as to identify patterns' grammatical meaning which can happen in application context for the sake of recognizing sentences more correctly. ", |
| "cite_spans": [ |
| { |
| "start": 181, |
| "end": 185, |
| "text": "[13]", |
| "ref_id": "BIBREF13" |
| }, |
| { |
| "start": 215, |
| "end": 218, |
| "text": "[9]", |
| "ref_id": "BIBREF9" |
| }, |
| { |
| "start": 221, |
| "end": 225, |
| "text": "[10]", |
| "ref_id": "BIBREF10" |
| }, |
| { |
| "start": 228, |
| "end": 232, |
| "text": "[15]", |
| "ref_id": "BIBREF15" |
| }, |
| { |
| "start": 235, |
| "end": 239, |
| "text": "[16]", |
| "ref_id": "BIBREF16" |
| }, |
| { |
| "start": 242, |
| "end": 246, |
| "text": "[18]", |
| "ref_id": "BIBREF18" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Voice recognizer component", |
| "sec_num": "3." |
| }, |
| { |
| "text": "Text result 7Speech 1Semantic tree (4) Text 3\u2022 Labelling, building dictionary.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "CENTRAL PROCESSOR", |
| "sec_num": null |
| }, |
| { |
| "text": "\u2022 Creating HMM prototypes for each of phone units.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "CENTRAL PROCESSOR", |
| "sec_num": null |
| }, |
| { |
| "text": "The training period's output is the file of trained HMM prototypes.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "CENTRAL PROCESSOR", |
| "sec_num": null |
| }, |
| { |
| "text": "\u2022 The file of trained HMM set is the result of the training period.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Recognizing period:", |
| "sec_num": "2." |
| }, |
| { |
| "text": "\u2022 Building dictionary.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Recognizing period:", |
| "sec_num": "2." |
| }, |
| { |
| "text": "\u2022 Extracting characteristics for voice series needing being recognized.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Recognizing period:", |
| "sec_num": "2." |
| }, |
| { |
| "text": "The recognizing period's output is the text series. Building a language prototypes involves the determining its grammar. The complexity of grammar depends on that of the system needing being recognized.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Recognizing period:", |
| "sec_num": "2." |
| }, |
| { |
| "text": "Grammar structure is a generalized graphic which implicates pattern sentences possibly occuring in application context. In our application, a part of the grammar file will be displayed as: Text-to-speech system includes two main phases: text analyzing (processing phase, standardizing text input in order to synthesize it) and text -to -speech phase (building speech signals from the former's results). The second phrase can be implemented by Formant text-to-speech [8] or Unitselection approach [1] , [8] \u2026 For Weather Voice, we have chosen unit-selection, the phrase is undergone in the next procedure: Figure 6 : Text-to-speech procedure by connecting to unitise. ", |
| "cite_spans": [ |
| { |
| "start": 466, |
| "end": 469, |
| "text": "[8]", |
| "ref_id": "BIBREF8" |
| }, |
| { |
| "start": 496, |
| "end": 499, |
| "text": "[1]", |
| "ref_id": "BIBREF0" |
| }, |
| { |
| "start": 502, |
| "end": 505, |
| "text": "[8]", |
| "ref_id": "BIBREF8" |
| } |
| ], |
| "ref_spans": [ |
| { |
| "start": 605, |
| "end": 613, |
| "text": "Figure 6", |
| "ref_id": null |
| } |
| ], |
| "eq_spans": [], |
| "section": "Recognizing period:", |
| "sec_num": "2." |
| }, |
| { |
| "text": "A. Query statement syntax.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Vietnamese language processing", |
| "sec_num": "4." |
| }, |
| { |
| "text": "Our system in all represents 18 query statement patterns as shown in the table 1. To represent query statement semantic, we use DCG [3] , [7] , [11] , [12] , [17] all nine representation patterns of query statement semantic are given in This pattern is the pattern number 1 in the table 3.", |
| "cite_spans": [ |
| { |
| "start": 132, |
| "end": 135, |
| "text": "[3]", |
| "ref_id": "BIBREF2" |
| }, |
| { |
| "start": 138, |
| "end": 141, |
| "text": "[7]", |
| "ref_id": "BIBREF7" |
| }, |
| { |
| "start": 144, |
| "end": 148, |
| "text": "[11]", |
| "ref_id": "BIBREF11" |
| }, |
| { |
| "start": 151, |
| "end": 155, |
| "text": "[12]", |
| "ref_id": "BIBREF12" |
| }, |
| { |
| "start": 158, |
| "end": 162, |
| "text": "[17]", |
| "ref_id": "BIBREF17" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Vietnamese language processing", |
| "sec_num": "4." |
| }, |
| { |
| "text": "We carry out transforming these semantic patterns into corresponding SQL statements in order to search corresponding weather information in the database. These data are automatically extracted from Yahoo Weather APL.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Vietnamese language processing", |
| "sec_num": "4." |
| }, |
| { |
| "text": "The experiments, at first, were conducted by each of system's components, which comprised voice recognizer, Vietnamese language processor and central processor. Afterwards, we experimented the entire system as well as performed surveys on users' reviews of the system including text-to-speech component.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "A. Speech recognizer component.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "The performance of speech recognizer is often evaluated by the metric WER (Word Error Rate), it is computed as the next formula:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "WER= (S+D+I) / N x 100%", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "Where:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "\u2022 N is the number of words,", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "\u2022 S is the number of substitutions,", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "\u2022 I is the number of insertions,", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "\u2022 D is the number of deletions.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "But here, we used the metric WAR (Word Accuracy Rate) in the evaluation of the system's performance by the formula:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "WAR= (1-(S+ D+ I) / N) x 100%", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "We gradually carried out the experiments classified in accordance to areas, sexes, ages, and trained participants. The system's accuracy is shown in the following tables 3, 4, 5 and 6. Table 4 : Based on sexes.", |
| "cite_spans": [], |
| "ref_spans": [ |
| { |
| "start": 185, |
| "end": 192, |
| "text": "Table 4", |
| "ref_id": null |
| } |
| ], |
| "eq_spans": [], |
| "section": "Experiments and evaluations", |
| "sec_num": "5." |
| }, |
| { |
| "text": "91% 94% Table 5 : Based on ages.", |
| "cite_spans": [], |
| "ref_spans": [ |
| { |
| "start": 8, |
| "end": 15, |
| "text": "Table 5", |
| "ref_id": null |
| } |
| ], |
| "eq_spans": [], |
| "section": "WAR Female Male", |
| "sec_num": null |
| }, |
| { |
| "text": "18-30 Others 95% 90% Table 6 : Based on trained participants and non-participants.", |
| "cite_spans": [], |
| "ref_spans": [ |
| { |
| "start": 21, |
| "end": 28, |
| "text": "Table 6", |
| "ref_id": null |
| } |
| ], |
| "eq_spans": [], |
| "section": "WAR", |
| "sec_num": null |
| }, |
| { |
| "text": "Non-participants", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Trained participants", |
| "sec_num": null |
| }, |
| { |
| "text": "B. Natural language processing component.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "Thanks to this component, we succeeded in experimenting 50 statements, the results for the statements involved exactly corresponded to our expectation. These statements are circumscribed within the syntastic structures having been built for the system. The system's ability to correctly process all these standard statements attests its stability and accuracy.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "The amplitude: For the statements which do not fall within the syntastic structures' circumscription, when returned by the system they will be assessed to be false. This evidence demonstrates our synstatic regulations DCG is incomplete and yet our grammar cannot cover every case which is likely to occur.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "If words grammar is added to it and synstatic regulations are improved, the system's amplitude will significantly expand.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "C. Surveys on users' reviews.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "During these surveys, users were asked the question coming from the system: \"Is the system easy to use?\". The answers are ranked in order of usefulness as mentioned in the table 7. D. Experimenting the entire system.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "The system is built in PC environment with Java language and SWI-Prolog version 6.6.5. ", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "98% 93%", |
| "sec_num": null |
| }, |
| { |
| "text": "The paper has offered the presentation of the Weather Voice system's structure and the approaches to build it. In the system, the Vietnamese language processing component, responsible for analyzing the statements' syntax and semantic meaning, assumes the central part.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Conclusion", |
| "sec_num": "6." |
| }, |
| { |
| "text": "According to our knowledge, this is one of the first systems in Vietnam to be equipped with effective natural language processing mechanism of voice application, this priority has made the system more intelligent and flexible. This research has also opened a new development to building and expanding question-answer systems which can understand Vietnamese and communicate with users in this language. In the future, we are also interested in developing voice applications for emotional analysis based on studies by Thien et al. [19, 20] .", |
| "cite_spans": [ |
| { |
| "start": 529, |
| "end": 533, |
| "text": "[19,", |
| "ref_id": "BIBREF19" |
| }, |
| { |
| "start": 534, |
| "end": 537, |
| "text": "20]", |
| "ref_id": "BIBREF20" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Conclusion", |
| "sec_num": "6." |
| } |
| ], |
| "back_matter": [], |
| "bib_entries": { |
| "BIBREF0": { |
| "ref_id": "b0", |
| "title": "Unit selection in a concatenative speech synthesis system using a large speech database", |
| "authors": [ |
| { |
| "first": "A", |
| "middle": [], |
| "last": "Hunt", |
| "suffix": "" |
| }, |
| { |
| "first": "W", |
| "middle": [], |
| "last": "Black", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Alan", |
| "suffix": "" |
| } |
| ], |
| "year": 1996, |
| "venue": "Pro c. ICASSP-96", |
| "volume": "1", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Hunt, A. Black and W. Alan, \"Unit selection in a concatenative speech synthesis system using a large speech database,\" Pro c. ICASSP-96, 1, pp. 373, 1996.", |
| "links": null |
| }, |
| "BIBREF1": { |
| "ref_id": "b1", |
| "title": "A Robust Vietnamese Voice Server for Automated Directory Assistance Application", |
| "authors": [ |
| { |
| "first": "Duong", |
| "middle": [], |
| "last": "Dau", |
| "suffix": "" |
| }, |
| { |
| "first": "Minh", |
| "middle": [], |
| "last": "Le", |
| "suffix": "" |
| }, |
| { |
| "first": "Cuong", |
| "middle": [], |
| "last": "Le", |
| "suffix": "" |
| }, |
| { |
| "first": "Quan", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2012, |
| "venue": "Viet Nam", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Duong Dau, Minh Le, Cuong Le and Quan Vu, \"A Robust Vietnamese Voice Server for Automated Directory Assistance Application,\" RIVF-VLSP 2012, Ho Chi Minh City, Viet Nam, 2012.", |
| "links": null |
| }, |
| "BIBREF2": { |
| "ref_id": "b2", |
| "title": "Prolog and Natural-Language Analysis", |
| "authors": [ |
| { |
| "first": "C", |
| "middle": [ |
| "N" |
| ], |
| "last": "Fernando", |
| "suffix": "" |
| }, |
| { |
| "first": "Stuart", |
| "middle": [ |
| "M" |
| ], |
| "last": "Pereira", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Shieber", |
| "suffix": "" |
| } |
| ], |
| "year": 2005, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "1--284", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Fernando C. N. Pereira and Stuart M. Shieber, Prolog and Natural-Language Analysis. Microtome Publishing, pp.1 -284, Massachusetts, 2005.", |
| "links": null |
| }, |
| "BIBREF3": { |
| "ref_id": "b3", |
| "title": "iSago: The Vietnamese Mobile Speech Assistant for Food-court and Restaurant Location", |
| "authors": [ |
| { |
| "first": "Hue", |
| "middle": [], |
| "last": "Nguyen", |
| "suffix": "" |
| }, |
| { |
| "first": "Truong", |
| "middle": [], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "Nhi", |
| "middle": [], |
| "last": "Le", |
| "suffix": "" |
| }, |
| { |
| "first": "Nhut", |
| "middle": [], |
| "last": "Pham", |
| "suffix": "" |
| }, |
| { |
| "first": "Quan", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2012, |
| "venue": "Viet Nam", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Hue Nguyen, Truong Tran, Nhi Le, Nhut Pham, Quan Vu, \"iSago: The Vietnamese Mobile Speech Assistant for Food-court and Restaurant Location,\" RIVF-VLSP 2012, Ho Chi Minh City, Viet Nam, 2012.", |
| "links": null |
| }, |
| "BIBREF4": { |
| "ref_id": "b4", |
| "title": "Windows NT 5.0 Brings You New Telephony Development Features with TAPI 3.0, Microsoft Systems Journal", |
| "authors": [ |
| { |
| "first": "Michelle", |
| "middle": [], |
| "last": "Quinton", |
| "suffix": "" |
| } |
| ], |
| "year": null, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Michelle Quinton, Windows NT 5.0 Brings You New Telephony Development Features with TAPI 3.0, Microsoft Systems Journal. [Online].", |
| "links": null |
| }, |
| "BIBREF5": { |
| "ref_id": "b5", |
| "title": "Available", |
| "authors": [], |
| "year": 1998, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Available: http://www.microsoft.com/msj/1198/tapi3/tapi3.aspx, 1998.", |
| "links": null |
| }, |
| "BIBREF6": { |
| "ref_id": "b6", |
| "title": "A Spoken Dialog System for Stock Information Inquiry", |
| "authors": [ |
| { |
| "first": "Nhut", |
| "middle": [], |
| "last": "Pham", |
| "suffix": "" |
| }, |
| { |
| "first": "Quan", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2012, |
| "venue": "Proc. IT@EDU", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Nhut Pham, Quan Vu, \"A Spoken Dialog System for Stock Information Inquiry,\" in Proc. IT@EDU, Ho Chi Minh City, Viet Nam, 2012.", |
| "links": null |
| }, |
| "BIBREF7": { |
| "ref_id": "b7", |
| "title": "Representation and Inference for Natural Language: A First Course in Computational Semantics", |
| "authors": [ |
| { |
| "first": "Patrick", |
| "middle": [], |
| "last": "Blackburn", |
| "suffix": "" |
| }, |
| { |
| "first": "Johan", |
| "middle": [], |
| "last": "Bos", |
| "suffix": "" |
| } |
| ], |
| "year": 2007, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "1--376", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Patrick Blackburn, Johan Bos, \"Representation and Inference for Natural Language: A First Course in Computational Semantics\". CSLI Press, pp. 1 -376, Chicago, 2007.", |
| "links": null |
| }, |
| "BIBREF8": { |
| "ref_id": "b8", |
| "title": "VOS: The Corpus-based Vietnamese Text-to-speech System", |
| "authors": [ |
| { |
| "first": "Quan", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2010, |
| "venue": "Journal on Information", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Quan Vu, \"VOS: The Corpus-based Vietnamese Text-to-speech System,\" Journal on Information, Technologies, anh Communications, 2010.", |
| "links": null |
| }, |
| "BIBREF9": { |
| "ref_id": "b9", |
| "title": "Nghi\u00ean c\u1ee9u x\u00e2y d\u1ef1ng h\u1ec7 th\u1ed1ng Voice Server v\u00e0 \u1ee9ng d\u1ee5ng cho c\u00e1c d\u1ecbch v\u1ee5 tr\u1ea3 l\u1eddi t\u1ef1 \u0111\u1ed9ng qua \u0111i\u1ec7n tho\u1ea1i", |
| "authors": [ |
| { |
| "first": "Quan", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2012, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Quan Vu et al., (2012). \"Nghi\u00ean c\u1ee9u x\u00e2y d\u1ef1ng h\u1ec7 th\u1ed1ng Voice Server v\u00e0 \u1ee9ng d\u1ee5ng cho c\u00e1c d\u1ecbch v\u1ee5 tr\u1ea3 l\u1eddi t\u1ef1 \u0111\u1ed9ng qua \u0111i\u1ec7n tho\u1ea1i\". Technical report, Research project, HCM City Department of Science and Technology, Viet Nam.", |
| "links": null |
| }, |
| "BIBREF10": { |
| "ref_id": "b10", |
| "title": "An upgrading SentiVoice-a system for querying hotel service reviews via phone", |
| "authors": [ |
| { |
| "first": "T", |
| "middle": [ |
| "K" |
| ], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "T", |
| "middle": [ |
| "T" |
| ], |
| "last": "Phan", |
| "suffix": "" |
| } |
| ], |
| "year": 2015, |
| "venue": "International Conference on Asian Language Processing (IALP)", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Tran, T.K., Phan, T.T.: An upgrading SentiVoice-a system for querying hotel service reviews via phone. 2015 International Conference on Asian Language Processing (IALP).", |
| "links": null |
| }, |
| "BIBREF11": { |
| "ref_id": "b11", |
| "title": "Formal Philosophy: Selected Papers of", |
| "authors": [ |
| { |
| "first": "Richard", |
| "middle": [], |
| "last": "Montague", |
| "suffix": "" |
| } |
| ], |
| "year": 1974, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "1--119", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Richard Montague, Formal Philosophy: Selected Papers of Richard Montague. Bell & Howell Information & Lea, pp. 1 -119, New Haven, 1974.", |
| "links": null |
| }, |
| "BIBREF12": { |
| "ref_id": "b12", |
| "title": "LING 364: Introduction to Formal Semantics", |
| "authors": [ |
| { |
| "first": "Sandiway", |
| "middle": [], |
| "last": "Fong", |
| "suffix": "" |
| } |
| ], |
| "year": null, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Sandiway Fong, \"LING 364: Introduction to Formal Semantics.", |
| "links": null |
| }, |
| "BIBREF13": { |
| "ref_id": "b13", |
| "title": "The HTK Book (version 3.4). Available: htk", |
| "authors": [ |
| { |
| "first": "Steve", |
| "middle": [], |
| "last": "Young", |
| "suffix": "" |
| } |
| ], |
| "year": 2006, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Steve Young et al, The HTK Book (version 3.4). Available: htk.eng.cam.ac.uk/docs/docs.shtml, 2006.", |
| "links": null |
| }, |
| "BIBREF14": { |
| "ref_id": "b14", |
| "title": "The Development of Vietnamese Corpora Toward Speech Translation System", |
| "authors": [ |
| { |
| "first": "Thang", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| }, |
| { |
| "first": "Mai", |
| "middle": [], |
| "last": "Luong", |
| "suffix": "" |
| } |
| ], |
| "year": 2012, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Thang Vu, Mai Luong, \"The Development of Vietnamese Corpora Toward Speech Translation System,\" RIVF-VLSP 2012, Ho Chi Minh City, Viet Nam, 2012.", |
| "links": null |
| }, |
| "BIBREF15": { |
| "ref_id": "b15", |
| "title": "Semantic Processing Mechanism for Listening and Comprehension in VNSCalendar System", |
| "authors": [], |
| "year": 2013, |
| "venue": "International Journal on Natural Language Computing (IJNLC)", |
| "volume": "2", |
| "issue": "2", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Thien Khai Tran, Dang Tuan Nguyen (2013). \"Semantic Processing Mechanism for Listening and Comprehension in VNSCalendar System\". International Journal on Natural Language Computing (IJNLC) Vol. 2, No.2, April 2013.", |
| "links": null |
| }, |
| "BIBREF16": { |
| "ref_id": "b16", |
| "title": "EDUVoice -a system for querying academic information via PSTN", |
| "authors": [ |
| { |
| "first": "", |
| "middle": [], |
| "last": "Tk Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "T", |
| "middle": [ |
| "A" |
| ], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Mai", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Nguyen", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Vu", |
| "suffix": "" |
| } |
| ], |
| "year": 2014, |
| "venue": "The Third Asian Conference on Information Systems", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "TK Tran, TCK Tran, TA Mai, NMH Nguyen, HT Vu. EDUVoice -a system for querying academic information via PSTN. The Third Asian Conference on Information Systems (ACIS 2014). Nha Trang.", |
| "links": null |
| }, |
| "BIBREF17": { |
| "ref_id": "b17", |
| "title": "An upgrading SentiVoice-a system for querying hotel service reviews via phone", |
| "authors": [ |
| { |
| "first": "T", |
| "middle": [ |
| "K" |
| ], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "T", |
| "middle": [ |
| "T" |
| ], |
| "last": "Phan", |
| "suffix": "" |
| } |
| ], |
| "year": null, |
| "venue": "2015 International Conference on", |
| "volume": "", |
| "issue": "", |
| "pages": "115--118", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Tran, T.K., Phan, T.T.: An upgrading SentiVoice-a system for querying hotel service reviews via phone. Asian Language Processing (IALP), 2015 International Conference on, 115-118.", |
| "links": null |
| }, |
| "BIBREF18": { |
| "ref_id": "b18", |
| "title": "Towards Building an Intelligent Call Routing System", |
| "authors": [ |
| { |
| "first": "", |
| "middle": [], |
| "last": "Tk Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Pham", |
| "suffix": "" |
| }, |
| { |
| "first": "", |
| "middle": [], |
| "last": "Van Huynh", |
| "suffix": "" |
| } |
| ], |
| "year": null, |
| "venue": "International Journal of Advanced Computer Science and Applications", |
| "volume": "7", |
| "issue": "1", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "TK Tran, DM Pham, B Van Huynh. Towards Building an Intelligent Call Routing System International Journal of Advanced Computer Science and Applications 7 (1).", |
| "links": null |
| }, |
| "BIBREF19": { |
| "ref_id": "b19", |
| "title": "A hybrid approach for building a Vietnamese sentiment dictionary", |
| "authors": [ |
| { |
| "first": "T", |
| "middle": [ |
| "K" |
| ], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "T", |
| "middle": [ |
| "T" |
| ], |
| "last": "Phan", |
| "suffix": "" |
| } |
| ], |
| "year": 2018, |
| "venue": "J. Intell. Fuzzy Syst", |
| "volume": "35", |
| "issue": "1", |
| "pages": "967--978", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Tran, T.K., Phan, T.T.: A hybrid approach for building a Vietnamese sentiment dictionary. J. Intell. Fuzzy Syst. 35(1), 967-978 (2018).", |
| "links": null |
| }, |
| "BIBREF20": { |
| "ref_id": "b20", |
| "title": "Mining opinion targets and opinion words from online reviews", |
| "authors": [ |
| { |
| "first": "T", |
| "middle": [ |
| "K" |
| ], |
| "last": "Tran", |
| "suffix": "" |
| }, |
| { |
| "first": "T", |
| "middle": [ |
| "T" |
| ], |
| "last": "Phan", |
| "suffix": "" |
| } |
| ], |
| "year": 2017, |
| "venue": "Int. J. Inf. Technol", |
| "volume": "9", |
| "issue": "3", |
| "pages": "239--249", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Tran, T.K., Phan, T.T.: Mining opinion targets and opinion words from online reviews. Int. J. Inf. Technol. 9(3), 239-249 (2017).", |
| "links": null |
| } |
| }, |
| "ref_entries": { |
| "FIGREF0": { |
| "uris": null, |
| "num": null, |
| "text": "The system: Hello. \u2212 The user: Please tell me what the weather is like in Can Tho tomorrow? \u2212 The system: Do you want to know about the weather of tomorrow in Can Tho? \u2212 The user: That's right. \u2212 -The system: The temperature is between 24 degrees and 34 tomorrow, it is sunny in the daytime, it rains in the evening and at night.", |
| "type_str": "figure" |
| }, |
| "FIGREF1": { |
| "uris": null, |
| "num": null, |
| "text": "achieving the above functions, the system must require the following components: \u2212 Voice recognizer component: to transform speech data which is human speech into data text. \u2212 Vietnamese language processing component: to analyze commands' syntax and semantic meaning from users. \u2212 Central processing component: to connect the other components each to another through such operations: Transforming data text from speech recognizer into standard data to be executed by Prolog in Vietnamese language processing. 2. Transforming commands' semantic expressions into a file of select statements which is sent to database and executing them.", |
| "type_str": "figure" |
| }, |
| "FIGREF2": { |
| "uris": null, |
| "num": null, |
| "text": "Filtering, arranging, and returning processed results via system to user. o Database: contains selected data. o Vietnamese language synthetizing component: transforms text data into speech.", |
| "type_str": "figure" |
| }, |
| "FIGREF3": { |
| "uris": null, |
| "num": null, |
| "text": "system structure.", |
| "type_str": "figure" |
| }, |
| "FIGREF4": { |
| "uris": null, |
| "num": null, |
| "text": ". Steps in building voice recognizer component. Building a voice recognizer component includes two main periods: 1. Training period:\u2022 Preparing voice data file needs training and codifying this file.", |
| "type_str": "figure" |
| }, |
| "FIGREF5": { |
| "uris": null, |
| "num": null, |
| "text": "B. Data to train. The file of data to train is recorded for 200 minutes in amount of 2,500 patterns. Those data must satisfy the criterion of 8,000 Hz, 16bit according to PCM format and be recorded in 50 different accents and in a quiet environment. The word list comprises 63 cities and towns' names over the country and key words concerning command sentences. C. Building language grammar. Our grammar is language prototypes providing information of sentences' syntax, semantics and word order. This component will help the system choose the best recognized results from the list of 'candidates' previously selected by the recognizing period. Sentence structures are likely to exist in application contexts.", |
| "type_str": "figure" |
| }, |
| "FIGREF6": { |
| "uris": null, |
| "num": null, |
| "text": "synthesizing (text-to-speech).", |
| "type_str": "figure" |
| }, |
| "TABREF0": { |
| "content": "<table/>", |
| "text": "transforming audible signals into visible text) and Voice synthesizing (transforming visible text into audible signals), their authors, Information Technology Institute (Vietnamese Science and Technology Institute) and Natural Science University (Vietnamese National University), have The 2018 Conference on Computational Linguistics and Speech Processing ROCLING 2018, pp. 126-139 \u00a9The Association for Computational Linguistics and Chinese Language Processing", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| }, |
| "TABREF2": { |
| "content": "<table><tr><td colspan=\"2\">Number Query statement pattern</td><td>Example</td></tr><tr><td>1</td><td>[weather] [place] [time]</td><td>What is the weather like in Hanoi today?</td></tr><tr><td>2</td><td>[weather] [place]</td><td>What is the weather like in Hanoi?</td></tr><tr><td>3</td><td>[weather] [time]</td><td>What is the weather like today?</td></tr><tr><td>4</td><td>[place] [time] [weather]</td><td>In Hanoi tomorrow, what is the weather like?</td></tr><tr><td>5</td><td>[place] [weather]</td><td>In HCM city what is the weather like?</td></tr><tr><td>6</td><td>[time] [weather]</td><td>After tomorrow what is the temperature?</td></tr><tr><td>7</td><td>[time] [place] [weather]</td><td>After tomorrow in Can Tho what is the weather like?</td></tr><tr><td>8</td><td>[temperature] [place] [time]</td><td>What is the temperature in Hanoi today?</td></tr><tr><td>9</td><td>[temperature] [place]</td><td>What is the temperature in Hanoi?</td></tr><tr><td>10</td><td>[temperature] [time]</td><td>What is the temperature today?</td></tr><tr><td>11</td><td>[place] [temperature] [time]</td><td>In Hanoi what is the temperature tomorrow?</td></tr><tr><td>12</td><td>[place] [temperature]</td><td>In HCM city what is the temperature?</td></tr><tr><td>13</td><td>[time] [temperature]</td><td>After tomorrow what is the temperature?</td></tr><tr><td>14</td><td>[time] [place] [temperature]</td><td>After tomorrow in Can Tho what is the temperature?</td></tr><tr><td>15</td><td>[place] [time] [state]</td><td>In Da Nang tomorrow does it rain?</td></tr><tr><td>16</td><td>[time] [place] [state]</td><td>Tomorrow in Da Nang is it hot?</td></tr><tr><td>17</td><td>[place] [state]</td><td>In Sai Gon does it rain?</td></tr><tr><td>18</td><td>[time] [state]</td><td>After tomorrow does it rain?</td></tr></table>", |
| "text": "Some query statements classified according to them.", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| }, |
| "TABREF3": { |
| "content": "<table><tr><td>Number</td><td>Meaning representations</td><td>Query statement patterns</td></tr><tr><td>1</td><td>query(weather, place, time)</td><td/></tr><tr><td>2</td><td>query (weather, place)</td><td>Structure patterns from 1 to 7 coming from table 2</td></tr><tr><td>3</td><td>query (weather, time)</td><td/></tr><tr><td>4</td><td>query (temperature, place,</td><td/></tr><tr><td/><td>time)</td><td/></tr><tr><td>5</td><td>query (temperature, place)</td><td>Structure patterns from 8 to 14 coming from table 2</td></tr><tr><td>6</td><td>query (temperature, time)</td><td/></tr><tr><td>7</td><td>yesno (place, state, time)</td><td/></tr><tr><td>8</td><td>yesno (place, state)</td><td>Structure patterns from 15 to 18 coming from table 2</td></tr><tr><td>9</td><td>yesno (state, time)</td><td/></tr><tr><td colspan=\"3\">Example 1: What is the weather like in Hanoi tomorrow?</td></tr><tr><td colspan=\"3\">DCG syntactic and semantic order will be defined as follows:</td></tr><tr><td colspan=\"3\">query(query(Weather,Place,Time)) --> n_weather(Weather),prep_place,n_place(Place),</td></tr><tr><td colspan=\"2\">n_time(Time),w_how.</td><td/></tr><tr><td colspan=\"2\">prep_place --> [in].</td><td/></tr><tr><td colspan=\"2\">n_weather(weather) --> [weather].</td><td/></tr><tr><td colspan=\"2\">n_place(place(c\u1ea7n_th\u01a1)) --> [c\u1ea7n_th\u01a1].</td><td/></tr><tr><td colspan=\"2\">n_time(time(tomorrow )--> [tomorrow].</td><td/></tr><tr><td colspan=\"2\">w_how --> [how].</td><td/></tr></table>", |
| "text": "Table 2. Representation patterns of query statement semantic.", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| }, |
| "TABREF4": { |
| "content": "<table><tr><td/><td>WAR</td><td/></tr><tr><td>North</td><td>Center</td><td>South</td></tr><tr><td>90%</td><td>88%</td><td>93%</td></tr></table>", |
| "text": "Based on areas.", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| }, |
| "TABREF5": { |
| "content": "<table><tr><td>Very useful</td><td>Quite useful</td><td>Slightly useful</td><td>Useless</td></tr><tr><td>25%</td><td>27%</td><td>23%</td><td>25%</td></tr></table>", |
| "text": "Indication of levels of the system's usefulness.", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| }, |
| "TABREF6": { |
| "content": "<table><tr><td>Number of query</td><td>100</td></tr><tr><td>statements</td><td/></tr><tr><td colspan=\"2\">Environment indoor</td></tr><tr><td colspan=\"2\">Sample rate 8kHz</td></tr><tr><td colspan=\"2\">Quantization 16bits</td></tr><tr><td colspan=\"2\">Format PCM</td></tr><tr><td colspan=\"2\">Device mobile phone</td></tr></table>", |
| "text": "Experiment parameter. The results given by the system are correct for 48 out of 50 Vietnamese query statements. As seen above, all the unexpected results happened during the identifying period. The feedback time is 2.6 seconds on average. E. Assessment Throughout the experiment, voice recognizer component has incorrectly identified 6 out of 50 experimented statements. But syntactically, among them 4 statements have always kept their initial meanings and then correctly been processed by natural language processing component; only the two left ones have received the wrong meanings because of the identifying period. By that, we have found that the important part is played by the natural language processing component, it can even rectify errors resulting from the identifying period.", |
| "num": null, |
| "type_str": "table", |
| "html": null |
| } |
| } |
| } |
| } |
