Abstract: This hearing aid comprises a microphone (101) and an external input terminal (102), a hearing aid processor (150) to which audio signals from the microphone (101) and the external input terminal (102) are inputted, and a receiver (113) to which audio signals that have undergone hearing aid processing by this hearing aid processor (150) are outputted. The hearing aid processor (150) has a mixer (112) that mixes audio signals from the microphone (101) with audio signals from the external input terminal (102) and outputs these audio signals to the receiver (113), a mix ratio determination unit (111) for determining the mix ratio between audio signals from the microphone (101) and audio signals from the external input terminal (102) in this mixer (112), and a facial movement detector (110) that is connected to this mix ratio determination unit (111).

Abstract: A speech recognition device is provided which includes: a language model storage unit which stores a language model indicating appearance probabilities of words or word sequences; an acoustic feature amount extracting unit and a checking unit which extract a feature amount of an inputted speech signal, and identifies the word or word sequence corresponding to the speech signal by checking the extracted feature amount with the language model stored in the language model storage unit; an obtained word signal receiving/analyzing unit which obtains and analyzes the word; and a language model adjusting unit which identifies the appearance probability of the word based on the time elapsed after obtaining the word by the obtained word signal receiving/analyzing unit and which adjusts the language model by reflecting the identified appearance probability on the language model stored in the language model storage unit.

Abstract: A voice output apparatus, enhancing a robustness of an interface between a user and the apparatus by transmitting, information to the user via text message and voice message. The voice output apparatus including a display unit (107) displaying a text message that is apparatus-transmitting information to be transmitted to the user, a delay unit (105), and a voice output unit (106) estimating a delay time necessary for an action taken by the user to visually identify the text message displayed by the display unit (107), and outputting, via voice message, the apparatus-transmitting information, when the delay time (T) passes after the text message is displayed.

Abstract: Provided are an operation assisting apparatus and a method thereof which accurately estimate an intention of a user and present a desired function more preferentially. The operation assisting apparatus estimates a function desired by a user based on a history of user operations for selecting options displayed on a menu screen having a hierarchical structure, and provides an operational assistance based on a result of the estimation.

Abstract: A speech recognition apparatus equipped with the garbage acoustic model storage unit storing the garbage acoustic model which learned the collection of unnecessary words. A feature value calculation unit calculates the feature parameter necessary for recognition by acoustically analyzing the unidentified input speech including the non-language speech per frame which is a unit for speech analysis. A garbage acoustic score calculation unit calculates the garbage acoustic score by comparing the feature parameter and the garbage acoustic model, and a garbage acoustic score correction unit corrects the garbage acoustic score calculated by the garbage acoustic score calculation unit so as to raise it in the frame where the non-language speech is inputted.

Abstract: A target estimation device that appropriately estimates a function targeted by a user from among functions of an operated apparatus includes: an item inputting unit (102) that obtains an item; a function representation storing unit (107) that stores function names of the operated apparatus; an identical word detecting unit (116) that detects a category of a word identical between each of the function names and the item; an operation intention calculating unit (117) that changes a method for calculating similarities between the item and the respective function names, for each of the function names, depending on the detected category, and that calculates the similarities as degrees of intentions of the user who has selected the item, using the changed calculation method; and a target estimation unit (112) that estimates a function so that the function having a higher degree of an intention is the function targeted by the user.

Abstract: A user support device which accurately judges that a user is in a confused state and supports the user is provided. The user support device includes: a confused state judging unit (107) which judges whether the user is in a confused state or not based on a tendency of variation in dwell time on which a decision making time until a next input action is executed is reflected in a sequence of input actions executed by the user who makes transitions between several menus arranged in a tree structure; and an operation support processing unit (109) which provides operation support when the confused state judging unit (107) makes a positive judgment that the user is in a confused state.

Abstract: A program guidance apparatus includes a recognition word storage unit (105) operable to store a past recognition word that is recognized by speech recognition in the past, a viewing history word storage unit (106) operable to store viewing history words that are the information of a viewed program and a dictionary creating unit (103) operable to create a customized recognition dictionary that is created by adding the past recognition word and viewing history words that are not included in the basic recognition dictionary to the basic recognition dictionary and another customized recognition dictionary to which weights are assigned using “item weight coefficient” according to the categories of words and “history weight coefficient” according to whether or not the word is recorded as a past recognition word or viewing history words.

Abstract: Provided is a speech recognition device which appropriately applies limitations on target words to be recognized which are obtained from outside of the speech recognition device, as well as to eliminate the uncomfortable feeling caused by the limitation processing.

Abstract: The present invention provides a speech recognition apparatus which appropriately performs speech recognition by generating, in real time, language models adapted to a new topic even in the case where topics are changed.

Abstract: To provide a speech recognition apparatus which appropriately performs speech recognition by generating, in real time, language models adapted to a new topic even in the case where topics are changed.

Abstract: The voice output apparatus, which enhances a robustness of an interface between a user and the apparatus by transmitting, information to the user via text message and voice message, is comprised of: a display unit (107) for displaying a text message that is apparatus-transmitting information to be transmitted to the user; and a delay unit (105) as well as a voice output unit (106) for estimating a delay time necessary for an action taken by the user to visually identify the text message displayed by the display unit (107), and outputting, via voice message, the apparatus-transmitting information, when the delay time (T) passes after the text message is displayed.

Abstract: The speech recognition apparatus (1) is equipped with the garbage acoustic model storage unit (110) storing the garbage acoustic model which learned the collection of the unnecessary words; the feature value calculation unit (101) which calculates the feature parameter necessary for recognition by acoustically analyzing the unidentified input speech including the non-language speech per frame which is a unit for speech analysis; the garbage acoustic score calculation unit (111) which calculates the garbage acoustic score by comparing the feature parameter and the garbage acoustic model; the garbage acoustic score correction unit (113) which corrects the garbage acoustic score calculated by the garbage acoustic score calculation unit (111) so as to raise it in the frame where the non-language speech is inputted; and the recognition result output unit (105) which outputs, as the recognition result of the unidentified input speech, the word string with the highest cumulative score of the language score, the word

Abstract: A language model generation and accumulation apparatus (10) that generates and accumulates language models for speech recognition is comprised of: a higher-level N-gram generation and accumulation unit (11) that generates and accumulates a higher-level N-gram language model obtained by modeling each of a plurality of texts as a string of words including a word string class having a specific linguistic property; and a lower class dependent word N-gram generation and accumulation unit (12) that generates and accumulates a lower-level N-gram language model obtained by modeling a sequence of words included in each word string class.

Abstract: A program guidance apparatus comprises a recognition word storage unit (105) operable to store a past recognition word that is recognized by speech recognition in the past, a viewing history word storage unit (106) operable to store viewing history words that are the information of a viewed program and a dictionary creating unit (103) operable to create a customized recognition dictionary that is created by adding the past recognition word and viewing history words that are not included in the basic recognition dictionary to the basic recognition dictionary and another customized recognition dictionary to which weights are assigned using “item weight coefficient” according to the categories of words and “history weight coefficient” according to whether or not the word is recorded as a past recognition word or viewing history words.

Abstract: A sentence or text based input utterance is prepared into word-string candidates each based on one to several words, starting from the beginning of a sentence. The candidates are displayed. A user is allowed to successively select and fix a candidate, to prepare a candidate for the following utterance on the basis of the selected candidate. The present invention is a text input method and apparatus adapted to repeat these processes. This eliminates the necessity to keep a memory space for search taking account of a number of combinations of word-string candidates, thus greatly reducing the storage capacity. Reduction is also possible in speech-recognition process amount. These makes possible apparatus size reduction. Furthermore, the user who is allowed to input one or more continuous utterances is free from troublesomeness as encountered in word-based input.

Abstract: An information retrieval system using an interactive interface is provided. Various factors, such as a frequency of user's use of the system and user's desires, are reflected in the progress of dialogs performed through the interface between a user and system. The dialogs are performed on a frame-driven technique. A task to be retrieved is provided by a frame in which a plurality of conditions are given as slots. A score determining significance of each slot is given to the slot. The score for each slot can be used in response to a user's manual operation or automatic search made into dialog control, degrees of user's leaning, contents to be retrieved, and others. The order and/or the number of times to make a user fill slots with conditions can be changed dynamically and shortened, thereby dialog control being efficient.