JP2006189730A5 - - Google Patents
Download PDFInfo
- Publication number
- JP2006189730A5 JP2006189730A5 JP2005003119A JP2005003119A JP2006189730A5 JP 2006189730 A5 JP2006189730 A5 JP 2006189730A5 JP 2005003119 A JP2005003119 A JP 2005003119A JP 2005003119 A JP2005003119 A JP 2005003119A JP 2006189730 A5 JP2006189730 A5 JP 2006189730A5
- Authority
- JP
- Japan
- Prior art keywords
- dialog
- recognition
- known degree
- vocabulary
- recognized
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000003993 interaction Effects 0.000 claims 14
- 230000002452 interceptive Effects 0.000 claims 7
Claims (15)
前記音声対話装置が、入力された音声の認識結果を出力する音声認識ステップと、
前記音声対話装置が有する認識語彙既知度合記憶部に記憶されている、前記複数の対話状態のそれぞれの状態において前記音声対話装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定ステップと、
前記音声認識ステップにおいて認識された前記認識結果と、前記認識語彙既知度合決定ステップにおいて決定された、現在の対話状態における認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定ステップと、
前記対話決定ステップにおいて決定された対話内容を出力する出力ステップと
を含むことを特徴とする音声対話方法。 A voice dialogue method by a voice dialogue device for dialogue with a user while presenting a plurality of dialogue states ,
A speech recognition step the spoken dialogue apparatus, for outputting a recognition result of the input speech,
A recognition vocabulary stored in the recognition vocabulary known degree storage unit of the voice interaction device and indicating how much the user understands the vocabulary that the speech interaction device can recognize in each of the plurality of dialogue states A recognition vocabulary known degree determination step for determining a recognized vocabulary known degree in the current dialog state using the known degree;
And the recognition result recognized in the voice recognition step was determined in the recognition vocabulary known degree determining step, and the recognition vocabulary known degree in the current dialog state based on the interaction in the next dialog state and the interactive state A dialog decision step to determine the content;
An output step for outputting the content of the dialog determined in the dialog determination step.
前記音声対話装置が、入力された音声を認識していないという認識結果であったとき、 When the voice interaction device has a recognition result that the input voice is not recognized,
前記現在の対話状態における認識語彙既知度合が、所定値を満たすかどうかを判定する認識語彙既知度合判定ステップと、 A recognized vocabulary known degree determination step for determining whether or not the recognized vocabulary known degree in the current dialog state satisfies a predetermined value;
前記認識語彙既知度合判定ステップにより、前記現在の対話状態における認識語彙既知度合が所定値を満たしていると判定されるときは、音声による再入力を促すことを決定し、前記現在の対話状態における認識語彙既知度合が所定値を満たしていないと判定されるときは、前記現在の対話状態における認識語彙既知度合に基づく対話を行うことを決定する対話状態決定ステップと When the recognized vocabulary known degree determination step determines that the recognized vocabulary known degree in the current conversation state satisfies a predetermined value, it is determined to prompt re-input by voice, and in the current conversation state A dialog state determining step for determining that a dialogue based on the recognized vocabulary known level in the current dialog state is to be performed when it is determined that the recognized vocabulary known level does not satisfy a predetermined value;
を含むことを特徴とする請求項1記載の音声対話方法。 The voice interaction method according to claim 1, further comprising:
対象の対話状態における入力モード毎の前記認識語彙既知度合をあらかじめ格納した既知度合テーブルを用いて、前記認識語彙既知度合を決定する
ことを特徴とする請求項1記載の音声対話方法。 In the recognition vocabulary known degree determination step,
The spoken dialogue method according to claim 1, wherein the recognized vocabulary known degree is determined using a known degree table in which the recognized vocabulary known degree for each input mode in a target dialogue state is stored in advance.
対象の対話状態における入力モード、認識語彙の変動に関する認識語彙変動情報、認識語彙の属性を示す認識語彙属性情報、全認識対象語彙数、表示認識対象語彙数、ユーザ自身の情報、ユーザのシステム使用履歴、対話進行状態、画面や応答音声による認識語彙に関する情報量の少なくとも一つを用いて、前記認識語彙既知度合を算出する
ことを特徴とする請求項1記載の音声対話方法。 In the recognition vocabulary known degree determination step,
Input mode in the conversation state of the target, recognition vocabulary fluctuation information regarding recognition vocabulary fluctuation, recognition vocabulary attribute information indicating the attributes of the recognition vocabulary, number of all recognition target vocabulary, number of display recognition target vocabulary, user's own information, user system use The spoken dialogue method according to claim 1, wherein the recognition vocabulary known degree is calculated using at least one of information relating to a recognized vocabulary based on a history, a dialogue progress state, a screen, and response voice.
前記出力ステップでは、前記対話決定ステップにおいて決定された前記対話の画面または音声応答の少なくとも1つを出力する
ことを特徴とする請求項1記載の音声対話方法。 In the dialog determination step, at least one of a dialog screen or a voice response is determined as the dialog content,
The voice dialog method according to claim 1, wherein at the output step, at least one of a screen or a voice response of the dialog determined in the dialog determination step is output.
前記出力ステップでは、前記対話決定ステップにより作成された前記認識語彙既知度合を示す表示または音声応答の少なくとも1つを出力する
ことを特徴とする請求項1記載の音声対話方法。 In the dialog determination step, at least one of a display or a voice response for indicating the recognized vocabulary known degree is created,
The voice dialog method according to claim 1, wherein at the output step, at least one of a display or a voice response indicating the recognized vocabulary known degree created by the dialog determination step is output.
ことを特徴とする請求項1記載の音声対話方法。 2. The speech dialogue method according to claim 1, wherein, in the dialogue determination step, whether or not to include an explanation related to a recognition target vocabulary in the speech recognition step is determined based on the recognition vocabulary known degree.
ことを特徴とする請求項1記載の音声対話方法。 In the dialog determination step, when the recognition result recognized in the voice recognition step is determined as an unknown word, the recognition is performed to determine whether the dialog content is a dialog content that prompts input again or a detailed dialog content. The speech dialogue method according to claim 1, wherein the speech dialogue method is determined based on a vocabulary known degree.
ことを特徴とする請求項8記載の音声対話方法。 9. The voice dialog method according to claim 8, wherein, in the dialog determination step, when the dialog content that prompts input again is determined, the voice recognition parameter in the voice recognition step is changed according to the number of re-inputs.
ことを特徴とする請求項8記載の音声対話方法。 9. The voice dialogue method according to claim 8, wherein in the dialogue determination step, when the detailed dialogue content is determined, the dialogue content is further changed based on the recognized vocabulary known degree.
前記情報検索装置が、入力された音声の認識結果を出力する音声認識ステップと、
前記情報検索装置が有する認識語彙既知度合記憶部に記憶されている、前記複数の対話状態のそれぞれの状態において前記情報検索装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定ステップと、
前記音声認識ステップにおいて認識された前記認識結果と、前記認識語彙既知度合決定ステップにおいて決定された、現在の対話状態における前記認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定ステップと、
前記対話決定ステップにおいて決定された対話内容を出力する出力ステップと、
前記出力ステップにおいて出力されている前記対話内容が情報検索を受け付ける内容である場合に、前記音声認識ステップにおいて認識された前記認識結果に基づいて情報を検索する情報検索ステップと
を含むことを特徴とする情報検索方法。 An information search method by an information search device for searching for information while presenting a plurality of dialog states ,
A speech recognition step wherein the information retrieval apparatus, for outputting a recognition result of the input speech,
A recognition vocabulary stored in the recognition vocabulary known degree storage unit of the information search device and indicating how much the user knows the vocabulary that the information search device can recognize in each of the plurality of dialog states A recognition vocabulary known degree determination step for determining a recognized vocabulary known degree in the current dialog state using the known degree;
And the recognition result recognized in the voice recognition step, the determined in the recognition vocabulary known degree determining step, and the recognition vocabulary known degree in the current dialog state, based on, in the next dialog state and the interactive state A dialog determination step for determining a dialog content;
An output step for outputting the content of the dialog determined in the dialog determination step;
An information search step of searching for information based on the recognition result recognized in the voice recognition step when the dialogue content output in the output step is a content for accepting an information search. How to search for information.
入力された音声の認識結果を出力する音声認識手段と、
前記複数の対話状態のそれぞれの状態において前記音声対話装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を記憶している認識語彙既知度合記憶手段と、
前記認識語彙既知度合記憶部に記憶されている前記認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定手段と、
前記音声認識手段で認識された前記認識結果と、前記認識語彙既知度合決定手段で決定された、現在の対話状態における認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定手段と、
前記対話決定手段で決定された対話内容を出力する出力手段と
を備えることを特徴とする音声対話装置。 A voice interaction device for interacting with a user while presenting a plurality of interaction states ,
A speech recognition means for outputting a recognition result of the input speech,
A recognized vocabulary known degree storage means for storing a recognized vocabulary known degree indicating how much the user knows a vocabulary that can be recognized by the voice interactive apparatus in each of the plurality of dialogue states;
A recognized vocabulary known degree determination means for determining a recognized vocabulary known degree in a current dialog state using the recognized vocabulary known degree stored in the recognized vocabulary known degree storage unit ;
And the recognition result recognized by the speech recognition means, as determined by the recognition vocabulary known degree determining means, and the recognition vocabulary known degree in the current dialog state based on the interaction in the next dialog state and the interactive state A dialog determination means for determining the content;
A voice dialog device comprising: output means for outputting the dialog content determined by the dialog determination means.
入力された音声の認識結果を出力する音声認識手段と、
前記複数の対話状態のそれぞれの状態において前記情報検索装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を記憶している認識語彙既知度合記憶手段と、
前記認識語彙既知度合記憶部に記憶されている前記認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定手段と、
前記音声認識手段で認識された前記認識結果と、前記認識語彙既知度合決定手段で決定された、現在の対話状態における前記認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定手段と、
前記対話決定手段で決定された対話内容を出力する出力手段と、
前記出力手段で出力されている前記対話内容が情報検索を受け付ける内容である場合に、前記音声認識手段で認識された前記認識結果に基づいて情報を検索する情報検索手段と
を備えることを特徴とする情報検索装置。 An information retrieval device for retrieving information while presenting a plurality of dialog states ,
A speech recognition means for outputting a recognition result of the input speech,
A recognized vocabulary known degree storage means for storing a recognized vocabulary known degree indicating how much a user knows a vocabulary that can be recognized by the information search device in each of the plurality of dialogue states;
A recognized vocabulary known degree determination means for determining a recognized vocabulary known degree in a current dialog state using the recognized vocabulary known degree stored in the recognized vocabulary known degree storage unit ;
And the recognition result recognized by the speech recognition means, the determined recognition vocabulary known degree determining means, and the recognition vocabulary known degree in the current dialog state, based on, in the next dialog state and the interactive state A dialog determining means for determining a dialog content;
Output means for outputting the content of the dialog determined by the dialog determination means;
An information search means for searching for information based on the recognition result recognized by the voice recognition means when the dialogue content output by the output means is a content for accepting an information search. Information retrieval device.
前記音声対話装置が、入力された音声の認識結果を出力する音声認識ステップと、
前記音声対話装置が有する認識語彙既知度合記憶部に記憶されている、前記複数の対話状態のそれぞれの状態において前記音声対話装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定ステップと、
前記音声認識ステップにおいて認識された前記認識結果と、前記認識語彙既知度合決定ステップにおいて決定された、現在の対話状態における認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定ステップと、
前記対話決定ステップにおいて決定された対話内容を出力する出力ステップとを前記音声対話装置に実行させる
ことを特徴とするプログラム。 A program for a voice interaction device that interacts with a user while presenting a plurality of interaction states ,
A speech recognition step the spoken dialogue apparatus, for outputting a recognition result of the input speech,
A recognition vocabulary stored in the recognition vocabulary known degree storage unit of the voice interaction device and indicating how much the user understands the vocabulary that the speech interaction device can recognize in each of the plurality of dialogue states A recognition vocabulary known degree determination step for determining a recognized vocabulary known degree in the current dialog state using the known degree;
And the recognition result recognized in the voice recognition step was determined in the recognition vocabulary known degree determining step, and the recognition vocabulary known degree in the current dialog state based on the interaction in the next dialog state and the interactive state A dialog decision step to determine the content;
A program for causing the voice interaction device to execute an output step for outputting the content of the dialogue determined in the dialogue determination step.
前記情報検索装置が、入力された音声の認識結果を出力する音声認識ステップと、
前記情報検索装置が有する認識語彙既知度合記憶部に記憶されている、前記複数の対話状態のそれぞれの状態において前記情報検索装置が認識できる語彙をユーザがどの程度把握しているのかを示す認識語彙既知度合を用いて、現在の対話状態における認識語彙既知度合を決定する認識語彙既知度合決定ステップと、
前記音声認識ステップにおいて認識された前記認識結果と、前記認識語彙既知度合決定ステップにおいて決定された、現在の対話状態における前記認識語彙既知度合と、に基づいて、次の対話状態および当該対話状態における対話内容を決定する対話決定ステップと、
前記対話決定ステップにおいて決定された対話内容を出力する出力ステップと、
前記出力ステップにおいて出力されている前記対話内容が情報検索を受け付ける内容である場合に、前記音声認識ステップにおいて認識された前記認識結果に基づいて情報を検索する情報検索ステップとを前記情報検索装置に実行させる
ことを特徴とするプログラム。 A program for an information retrieval device that retrieves information while presenting a plurality of dialog states ,
A speech recognition step wherein the information retrieval apparatus, for outputting a recognition result of the input speech,
A recognition vocabulary stored in the recognition vocabulary known degree storage unit of the information search device and indicating how much the user knows the vocabulary that the information search device can recognize in each of the plurality of dialog states A recognition vocabulary known degree determination step for determining a recognized vocabulary known degree in the current dialog state using the known degree;
And the recognition result recognized in the voice recognition step, the determined in the recognition vocabulary known degree determining step, and the recognition vocabulary known degree in the current dialog state, based on, in the next dialog state and the interactive state A dialog determination step for determining a dialog content;
An output step for outputting the content of the dialog determined in the dialog determination step;
When the dialogue content being output in said output step is the content that accepts information search, and information retrieval step of retrieving information on the basis of the recognized the recognition result in the speech recognition step to the information retrieval device A program characterized by being executed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2005003119A JP4634156B2 (en) | 2005-01-07 | 2005-01-07 | Voice dialogue method and voice dialogue apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2005003119A JP4634156B2 (en) | 2005-01-07 | 2005-01-07 | Voice dialogue method and voice dialogue apparatus |
Publications (3)
Publication Number | Publication Date |
---|---|
JP2006189730A JP2006189730A (en) | 2006-07-20 |
JP2006189730A5 true JP2006189730A5 (en) | 2008-02-14 |
JP4634156B2 JP4634156B2 (en) | 2011-02-16 |
Family
ID=36796996
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2005003119A Expired - Fee Related JP4634156B2 (en) | 2005-01-07 | 2005-01-07 | Voice dialogue method and voice dialogue apparatus |
Country Status (1)
Country | Link |
---|---|
JP (1) | JP4634156B2 (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5137853B2 (en) * | 2006-12-28 | 2013-02-06 | 三菱電機株式会社 | In-vehicle speech recognition device |
JP4915665B2 (en) * | 2007-04-18 | 2012-04-11 | パナソニック株式会社 | Controller with voice recognition function |
JP2013092823A (en) * | 2011-10-24 | 2013-05-16 | Nifty Corp | Information processing unit, program, and information retrieval system |
JP2016206960A (en) * | 2015-04-23 | 2016-12-08 | 日本電信電話株式会社 | Voice video input/output device |
JP2017167366A (en) * | 2016-03-16 | 2017-09-21 | Kddi株式会社 | Communication terminal, communication method, and program |
JP6628853B2 (en) * | 2018-10-09 | 2020-01-15 | 日本電信電話株式会社 | Audio-video tracking device |
CN110450789B (en) * | 2019-08-13 | 2020-12-15 | 广州小鹏汽车科技有限公司 | Information processing method and device |
CN112652301B (en) * | 2019-10-12 | 2023-05-12 | 阿里巴巴集团控股有限公司 | Voice processing method, distributed system, voice interaction device and voice interaction method |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001296890A (en) * | 2000-04-12 | 2001-10-26 | Auto Network Gijutsu Kenkyusho:Kk | On-vehicle equipment handling proficiency discrimination device and on-vehicle voice outputting device |
JP2003177788A (en) * | 2001-12-12 | 2003-06-27 | Fujitsu Ltd | Audio interactive system and its method |
JP4223832B2 (en) * | 2003-02-25 | 2009-02-12 | 富士通株式会社 | Adaptive spoken dialogue system and method |
JP4166616B2 (en) * | 2003-04-21 | 2008-10-15 | 松下電器産業株式会社 | Preference information type data retrieval device |
JP2004333543A (en) * | 2003-04-30 | 2004-11-25 | Matsushita Electric Ind Co Ltd | System and method for speech interaction |
-
2005
- 2005-01-07 JP JP2005003119A patent/JP4634156B2/en not_active Expired - Fee Related
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11817085B2 (en) | Automatically determining language for speech recognition of spoken utterance received via an automated assistant interface | |
US11398236B2 (en) | Intent-specific automatic speech recognition result generation | |
JP2006189730A5 (en) | ||
US10037758B2 (en) | Device and method for understanding user intent | |
Schalkwyk et al. | “Your word is my command”: Google search by voice: A case study | |
EP3640938B1 (en) | Incremental speech input interface with real time feedback | |
US8731927B2 (en) | Speech recognition on large lists using fragments | |
US8543407B1 (en) | Speech interface system and method for control and interaction with applications on a computing system | |
JP5111607B2 (en) | Computer-implemented method and apparatus for interacting with a user via a voice-based user interface | |
WO2018000278A1 (en) | Context sensitive multi-round dialogue management system and method based on state machines | |
JP2000315096A5 (en) | ||
US9298811B2 (en) | Automated confirmation and disambiguation modules in voice applications | |
JP2001034293A (en) | Method and device for transferring voice | |
US8909528B2 (en) | Method and system for prompt construction for selection from a list of acoustically confusable items in spoken dialog systems | |
US9922650B1 (en) | Intent-specific automatic speech recognition result generation | |
EP3593346A1 (en) | Graphical data selection and presentation of digital content | |
JP2016001242A (en) | Question sentence creation method, device, and program | |
CN112346697A (en) | Method, device and storage medium for controlling equipment | |
JP4634156B2 (en) | Voice dialogue method and voice dialogue apparatus | |
JP2006243673A5 (en) | ||
US10621282B1 (en) | Accelerating agent performance in a natural language processing system | |
US10140981B1 (en) | Dynamic arc weights in speech recognition models | |
JP6772916B2 (en) | Dialogue device and dialogue method | |
JP6746886B2 (en) | Learning support device and program for the learning support device | |
Kessens et al. | A bottom-up method for obtaining information about pronunciation variation |