269 results about "Speech generation" patented technology

The present invention allows subscribers to an online information service to participate in real-time conferencing or chat sessions in which a message originating from a subscriber in accordance with a first language is translated to one or more languages before it is broadcast to the other conference areas. Messages in a first language are translated automatically to one or more other languages through language translation capabilities resident at online information service host computers. Access software that subscribers use for participating in conference is integrated with speech recognition and speech generation software such that a subscriber may speak the message he or she would like to share with other participants and may hear the messages from the other participants in the conference. Speech-to-speech translation may be accomplished as a message spoken into a computer microphone in accordance with a first language may be recited by a remote computer in accordance with a second language.

A system and method are provided for improved speech synthesis, wherein text data is pre-processed according to updated grammar rules or a selected group of grammar rules. In one embodiment, the TTS system comprises a first memory adapted to store a text information database, a second memory adapted to store grammar rules, and a receiver adapted to receive update data regarding the grammar rules. The system also includes a TTS engine adapted to retrieve at least one text entry from the text information database, pre-process the at least one text entry by applying the updated grammar rules to the at least one text entry, and generate speech based at least in part on the least one pre-processed text entry.

The present invention discloses a solution for customizing synthetic voice characteristics in a user specific fashion. The solution can establish a communication between a user and a voice response system. A data store can be searched for a speech profile associated with the user. When a speech profile is found, a set of speech output characteristics established for the user from the profile can be determined. Parameters and settings of a text-to-speech engine can be adjusted in accordance with the determined set of speech output characteristics. During the established communication, synthetic speech can be generated using the adjusted text-to-speech engine. Thus, each detected user can hear a synthetic speech generated by a different voice specifically selected for that user. When no user profile is detected, a default voice or a voice based upon a user's speech or communication details can be used.

In several embodiments, a speech generation device is disclosed. The speech generation device may generally include a projector configured to project images in the form of a projected display onto a projection Surface, an optical input device configured to detect an input directed towards the projected display and a speaker configured to generate an audio output. In addition, the speech generation device may include a processing unit communicatively coupled to the projector, the optical input device and the speaker. The processing unit may include a processor and related computer readable medium configured to store instructions executable by the processor, wherein the instructions stored on the computer readable medium configure the speech generation device to generate text-to-speech output.

A method and an apparatus for improved duration modeling of phonemes in a speech synthesis system are provided. According to one aspect, text is received into a processor of a speech synthesis system. The received text is processed using a sum-of-products phoneme duration model that is used in either the formant method or the concatenative method of speech generation. The phoneme duration model, which is used along with a phoneme pitch model, is produced by developing a non-exponential functional transformation form for use with a generalized additive model. The non-exponential functional transformation form comprises a root sinusoidal transformation that is controlled in response to a minimum phoneme duration and a maximum phoneme duration. The minimum and maximum phoneme durations are observed in training data. The received text is processed by specifying at least one of a number of contextual factors for the generalized additive model. An inverse of the non-exponential functional transformation is applied to duration observations, or training data. Coefficients are generated for use with the generalized additive model. The generalized additive model comprising the coefficients is applied to at least one phoneme of the received text resulting in the generation of at least one phoneme having a duration. An acoustic sequence is generated comprising speech signals that are representative of the received text.

Text-to-speech (TTS) generation is used in conjunction with large vocabulary speech recognition to say words selected by the speech recognition. The software for performing the large vocabulary speech recognition can share speech modeling data with the TTS software. TTS or recorded audio can be used to automatically say both recognized text and the names of recognized commands after their recognition. The TTS can automatically repeats text recognized by the speech recognition after each of a succession of end of utterance detections. A user can move a cursor back or forward in recognized text, and the TTS can speak one or more words at the cursor location after each such move. The speech recognition can be used to produces a choice list of possible recognition candidates and the TTS can be used to provide spoken output of one or more of the candidates on the choice list.

Techniques are disclosed for automatically generating structured documents based on speech, including identification of relevant concepts and their interpretation. In one embodiment, a structured document generator uses an integrated process to generate a structured textual document (such as a structured textual medical report) based on a spoken audio stream. The spoken audio stream may be recognized using a language model which includes a plurality of sub-models arranged in a hierarchical structure. Each of the sub-models may correspond to a concept that is expected to appear in the spoken audio stream. Different portions of the spoken audio stream may be recognized using different sub-models. The resulting structured textual document may have a hierarchical structure that corresponds to the hierarchical structure of the language sub-models that were used to generate the structured textual document.

The invention discloses a conservational robot system. A human face tracking and recognizing module tracks and recognizes a human face image captured by a camera; a human facial expression recognizing module recognizes the expression; and semantic meaning is recognized after voice signals pass through a voice picking module and a voice recognizing module. The robot system understands human demands according to facial expressions and/or voice and then forms conservation statement through a knowledge inquiry module and generates voice for communication with humans through a voice generating module. The conservational robot system has voice recognizing and understanding abilities and can understand commands of users. The invention can be applied to schools, families, hotels, companies, airports, bus stations, docks, meeting rooms and so on for education, chat, conservation, consultation, etc. In addition, the invention can also help users with propaganda and introduction, guest reception, business inquiry, secretary service, foreign language interpretation, etc.

A method for retrieving information includes detecting a first neural activity of a user, wherein the first neural activity corresponds to a key thought of the user; in response to detecting the first neural activity, generating a search query based on speech occurring prior to detecting the first neural activity; retrieving information based on the search query; and transmitting the information to one or more output devices.

The invention discloses a voice production method based on a deep convolutional generative adversarial network. The steps comprise: (1) acquiring voice signal samples; (2) preprocessing the voice signal samples; (3) inputting the voice signal samples to a deep convolutional generative adversarial network; (4) training the input voice signals; (5) generating voice signals similar to real voice contents. A tensorflow is used as a learning framework, and a deep convolutional generative adversarial network algorithm is used to train large quantity of voice signals. A dynamic game process of a distinguishing network D and a generation network G in the deep convolutional generative adversarial network is used to finally generate a natural voice signal close to original learning contents. The method generates voice based on the deep convolutional generative adversarial network, and solves problems that an intelligent device is overly dependent on a fixed voice library to sound in a man-machine face-to-face communication process, and mode is monotonous and is lack of variations and is not natural enough.

An embedded device for playing media files is capable of generating a play list of media files based on input speech from a user. It includes an indexer generating a plurality of speech recognition grammars. According to one aspect of the invention, the indexer generates speech recognition grammars based on contents of a media file header of the media file. According to another aspect of the invention, the indexer generates speech recognition grammars based on categories in a file path for retrieving the media file to a user location. When a speech recognizer receives an input speech from a user while in a selection mode, a media file selector compares the input speech received while in the selection mode to the plurality of speech recognition grammars, thereby selecting the media file.

A speech generation device is disclosed. The speech generation device may include a head mounted display unit having a variety of different components that enhance the functionality of the speech generation device. The speech generation device may further include computer-readable medium that, when executed by a processor, instruct the speech generation device to perform desired functions.

Systems and methods of providing electronic features for creating a customized media player interface for an electronic device include providing a graphical user interface design area having a plurality of display elements. Electronic input signals then may define for association with selected display elements one or more electronic actions relative to the initiation and control of media files accessible by the electronic device (e.g., playing, pausing, stopping, adjusting play speed, adjusting volume, adjusting current file position, toggling modes such as repeat or shuffle, and/or establishing, viewing and/or clearing a playlist.) Additional electronic input signals may define for association with selected display elements labels such as action identification labels or media status labels. A graphical user interface is then initiated on an electronic display associated with an electronic device, wherein the graphical user interface comprises media player features corresponding to the plurality of display elements and associated electronic actions and/or labels.

The data control unit in the voice synthesizing server of the present invention receives from a client data to be processed to synthesize voice, and determines the type of the received data. If the received data are text data, they are outputted to the pronunciation symbol generating unit to generate pronunciation symbols. If the received data are pronunciation symbols, they are outputted to the acoustic parameter generating unit to generate acoustic parameters. If the received data are acoustic parameters, they are outputted to the waveform generating unit to generate voice waveforms, and the generated voice waveforms are sent from the data sending unit to the client. The client receives the voice waveforms from the voice generating server, and outputs the received voice waveforms as voice from the voice output unit.

A method for speech recognition can include generating a context-enhanced database from a system input. A voice-generated output can be generated from a speech signal by performing a speech recognition task to convert the speech signal into computer-processable segments. During the speech recognition task, the context-enhanced database can be accessed to improve the speech recognition rate. Accordingly, the speech signal can be interpreted with respect to words included within the context-enhanced database. Additionally, a user can edit or correct an output in order to generate the final voice-generated output which can be made available.

The invention discloses a television voice voting method, a television voice voting system and a television voice voting terminal. The television voice voting method comprise the following steps: a vote event is formed by the voting system and sent to a digital television terminal, wherein the vote event comprises a vote event identification (ID), vote content information and validity period information; the vote event is received by the digital television terminal and the vote event within the period of validity is showed; voice input is received, voice is identified, and voice voting information is generated and sent to the voting system; a voting result is updated by the voting system according to the voice voting information, the voting result is sent to the digital television terminal and showed. Due to the fact that the digital television terminal receives the vote event which is sent by the voting system, identifies the voice voting information of a user, sends a vote which is chosen by the user to the voting system, receives and shows the updated voting result which is sent back by the system, the function of voice television voting is achieved. The method through which users participate in a television voting activity is more convenient, faster and more interesting.

The present invention relates to systems and methods for enhancing the holistic and temporal speech perception processes of a learning-impaired subject. A subject listens to a sound stimulus which induces the perception of verbal transformations. The subject records the verbal transformations which are then used to create further sound stimuli in the form of semantic-like phrases and an imaginary story. Exposure to the sound stimuli enhances holistic speech perception of the subject with cross-modal benefits to speech production, reading and writing. The present invention has application to a wide range of impairments including, Specific Language Impairment, language learning disabilities, dyslexia, autism, dementia and Alzheimer's.

A device and method for pass-phrase modeling for speaker verification and a speaker verification system are provided. The device comprises a front end which receives enrollment speech from a target speaker, and a template generation unit which generates a pass-phrase template with a general speaker model based on the enrollment speech. With the device, method and system of the present disclosure, by taking the rich variations contained in a general speaker model into account, the robust pass-phrase modeling is ensured even the enrollment data is insufficient, even just one pass-phrase is available from a target speaker.

The invention discloses a voice generation method based on a generative adversarial network. According to the method, randomly-generated noise data meeting Gaussian distribution is converted into a simulation sample through a generative model; as the simulation sample does not have the language content, when the generative model and a discrimination model are circularly updated, generative capacities required to be learned by the generative model and discrimination capacities required to be learned by the discrimination model are correspondingly increased, and accordingly the generative capacities of the generative model and the discrimination capacities of the discrimination model are improved; when a contrast value between a training sample and the simulation sample is smaller than or equal to a preset threshold value, it is thought that the generative model has the capacity of generating real data; a voice database generated through the generative model has enough reality, and the recognition rate can be increased when the generative model is applied to identity recognition. Correspondingly, the voice generation method, a voice generation device and voice generation equipment based on the generative adversarial network and a computer readable storage medium have the same advantages.

Systems and methods of providing electronic features for creating context-aware vocabulary suggestions for an electronic device include providing a graphical user interface design area having a plurality of display elements. An electronic device user may be provided automated context-aware analysis from information from plural sources including GPS, compass, speaker identification (i.e., voice recognition), facial identification, speech content determination, user specifications, speech output monitoring, and software navigation monitoring to provide a selectable display of suggested vocabulary, previously stored words and phrases, or a keyboard as input options to create messages for text display and/or speech generation. The user may, optionally, manually specify a context.

An expressive speech-to-speech generation system and method which can generate expressive speech output by using expressive parameters extracted from the original speech signal to drive the standard TTS system. The system comprises: speech recognition means, machine translation means, text-to-speech generation means, expressive parameter detection means for extracting expressive parameters from the speech of language A, and expressive parameter mapping means for mapping the expressive parameters extracted by the expressive parameter detection means from language A to language B, and driving the text-to-speech generation means by the mapping results to synthesize expressive speech. The system and method can improve the quality of the speech output of the translating system or TTS system.

In a particular aspect, an apparatus includes an audio sensor configured to receive an input audio signal. The apparatus also includes speech generative circuitry configured to generate a synthesized audio signal based at least partly on automatic speech recognition (ASR) data associated with the input audio signal and based on one or more parameters indicative of state information associated with the input audio signal.