Abstract: The present disclosure provides an information processing method, including: acquiring a voice test signal, wherein the voice test signal is used for testing an influence parameter of a specified environment on a voice signal; acquiring a voice feedback signal obtained after the voice test signal is reflected by at least one reflecting object, wherein the at least one reflecting object is located in the specified environment; and estimating the influence parameter of the specified environment on the voice signal based on the voice test signal and the voice feedback signal acquired. In addition, the present disclosure also provides an information processing system, a computer system and a computer readable medium.

Abstract: A speech synthesizer includes a memory configured to store a plurality of sentences and prior information of a word classified into a minor class among a plurality of classes with respect to each sentence, and a processor configured to determine an oversampling rate of the word based on the prior information, determine the number of times of oversampling of the word using the determined oversampling rate and generate sentences including the word by the determined number of times of oversampling. The plurality of classes includes a first class corresponding to first reading break, a second class corresponding to second reading break greater than the first break and a third class corresponding to third reading break greater than the second break, and the minor class has a smallest count among the first to third classes in one sentence.

Abstract: Systems, apparatuses, and methods are described for controlling source tracking and delaying beamforming in a microphone array system. A source tracker may continuously determine a direction of an audio source. A source tracker controller may pause the source tracking of the source tracker if a user may continue to speak to the system. The source tracker controller may resume the source tracking of the source tracker if the user may cease to speak to the system, or when one or more pause durations have been reached.

Abstract: According to some aspects, a system for detecting a designated wake-up word is provided, the system comprising a plurality of microphones to detect acoustic information from a physical space having a plurality of acoustic zones, at least one processor configured to receive a first acoustic signal representing the acoustic information received by the plurality of microphones, process the first acoustic signal to identify content of the first acoustic signal originating from each of the plurality of acoustic zones, provide a plurality of second acoustic signals, each of the plurality of second acoustic signals substantially corresponding to the content identified as originating from a respective one of the plurality of acoustic zones, and performing automatic speech recognition on each of the plurality of second acoustic signals to determine whether the designated wake-up word was spoken.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for cross-lingual speech recognition are disclosed. In one aspect, a method includes the actions of determining a context of a second computing device. The actions further include identifying, by a first computing device, an additional pronunciation for a term of multiple terms. The actions further include including the additional pronunciation for the term in the lexicon. The actions further include receiving audio data of an utterance. The actions further include generating a transcription of the utterance by using the lexicon that includes the multiple terms and the pronunciation for each of the multiple terms and the additional pronunciation for the term. The actions further include after generating the transcription of the utterance, removing the additional pronunciation for the term from the lexicon. The actions further include providing, for output, the transcription.

Abstract: An augmented reality (AR) device can be configured to monitor ambient audio data. The AR device can detect speech in the ambient audio data, convert the detected speech into text, or detect keywords such as rare words in the speech. When a rare word is detected, the AR device can retrieve auxiliary information (e.g., a definition) related to the rare word from a public or private source. The AR device can display the auxiliary information for a user to help the user better understand the speech. The AR device may perform translation of foreign speech, may display text (or the translation) of a speaker's speech to the user, or display statistical or other information associated with the speech.

Abstract: According to an embodiment of the present disclosure, a method of updating a speech recognition model using a mobile agent in real-time comprises obtaining, in real-time, space type information for a particular space where the mobile agent is located, varying, in real-time, parameters of a speech recognition model used in the particular space based on the space type information, and performing a speech recognition service based on the speech recognition model including the varied parameters. Embodiments of the present disclosure may be related to artificial intelligence (AI) devices, unmanned aerial vehicles (UAVs), robots, augmented reality (AR) devices, virtual reality (VR) devices, and 5G service-related devices.

Abstract: A system and method are presented for dialogue tree generation. The dialogue tree may be used for generating a chatbot. Similar phrases from phrases comprising the interactions between a first party and a second party are group together from the first party of a cluster. For each group of similar phrases, percentages are determined and compared against a threshold occurrence rate. Anchors are generated and used in alignment in the determination of dialogue flows. Topic-specific dialogue trees may be determined from the dialogue flows. The topic-specific dialogue trees may be modified to generate a deterministic dialogue tree.

Abstract: A system and method are presented for configuring topic-specific chatbots. Clustering interaction transcripts between customers and agents of a contact center is performed to generated a plurality of interaction clusters. The clusters corresponding a topic. Topic-specific dialogue trees are extracted for each cluster. The trees comprise nodes connected by edges. The topic-specific dialogue tree is modified to generate a deterministic dialogue tree. The deterministic dialogue tree is used to configure a topic-specific chatbot to generate and automatically respond to messages regarding the topic.

Abstract: Provided are sentence embedding method and apparatus based on subword embedding and skip-thoughts. To integrate skip-thought sentence embedding learning methodology with a subword embedding technique, a skip-thought sentence embedding learning method based on subword embedding and methodology for simultaneously learning subword embedding learning and skip-thought sentence embedding learning, that is, multitask learning methodology, are provided as methodology for applying intra-sentence contextual information to subword embedding in the case of subword embedding learning. This makes it possible to apply a sentence embedding approach to agglutinative languages such as Korean in a bag-of-words form. Also, skip-thought sentence embedding learning methodology is integrated with a subword embedding technique such that intra-sentence contextual information can be used in the case of subword embedding learning.

Abstract: Embodiments of the present disclosure provide a method, an apparatus, a device and a storage medium for evaluating quality of an answer. The method includes extracting a question feature expression of a question and an answer feature expression of an answer with respect to the question, the question and the answer being represented in a form of text; determining a measurement of textual quality of the answer based on the answer feature expression; determining a measurement of correlation on semantics between the question and the answer based on the question feature expression and the answer feature expression; and determining a quality score of the answer with respect the question based on the measurement of textual quality and the measurement of correlations. Therefore, a high-quality answer may be accurately obtained.

Abstract: An interactive voice adapter for adaptive voice routing may establish a real-time communication session between a voice communication client and a text communication client and the voice adapter may receive the audio stream and the text information. The voice adapter may obtain adapted natural language text corresponding to the natural language audio by selectively accessing a speech-to-text service based on a selection criteria. The voice adapter may obtain adapted natural language audio corresponding to the natural language text by selectively accessing a text-to-speech service based on the selection criteria. The voice adapter may communicate the adapted natural language text to the text communication client and the adapted natural language audio to the voice communication client.

Abstract: The embodiments of the present disclosure provides a smart service method, apparatus and device, the method includes: a first device plays response information corresponding to an awaking word after receiving the awaking word; when the first device determines that a second device in an awake state exists in a device group that the first device is in, the first device transmits operation information of the first device to the second device, and receives operation information of the second device transmitted by the second device; the first device determines, according to the operation information of the first device and the operation information of the second device a target device from the first device and the second device, where the target device is a device in the device group to provide a smart service. The service resources are economized.

Abstract: An inter-channel phase difference (IPD) parameter extraction method and apparatus, where the extraction method includes obtaining a parameter obtaining an information extraction manner for a current frame of a multi-channel signal, obtaining an IPD parameter extraction manner for the current frame based on the parameter obtaining the information extraction manner, where the obtained IPD parameter extraction manner is one of at least two preset IPD parameter extraction manners, and obtaining an IPD parameter of the current frame based on the obtained IPD parameter extraction manner for the current frame.

Abstract: An audio decoder for providing a decoded audio information on the basis of an encoded audio information is configured to obtain decoded spectral values on the basis of an encoded information representing the spectral values. The audio decoder is configured to jointly decode two or more most significant bits per spectral value on the basis of respective symbol codes for a set of spectral values using an arithmetic decoding, wherein a respective symbol code represents two or more most significant bits per spectral value for one or more spectral values. The audio decoder is configured to decode one or more least significant bits associated with one or more of the spectral values in dependence on how much least significant bit information is available, such that one or more least significant bits associated with one or more of the spectral values are decoded.

Abstract: An audio decoder for providing a decoded audio information on the basis of an encoded audio information is configured to obtain decoded spectral values on the basis of an encoded information representing the spectral values. The audio decoder is configured to jointly decode two or more most significant bits per spectral value on the basis of respective symbol codes for a set of spectral values using an arithmetic decoding, wherein a respective symbol code represents two or more most significant bits per spectral value for one or more spectral values. The audio decoder is configured to decode one or more least significant bits associated with one or more of the spectral values in dependence on how much least significant bit information is available, such that one or more least significant bits associated with one or more of the spectral values are decoded.

Abstract: A mechanism is provided to implement suggestion of new entity types with discriminative importance analysis. The mechanism obtains a list of predefined intents from a chatbot designer. The mechanism receives an input sentence having a target intent within the list of predefined intents. The mechanism performs intent-specific importance analysis on the input sentence to generate an importance score for each token in the input sentence. The mechanism ranks the tokens in the input sentence by importance score and outputs a token with a highest importance score as a candidate entity type.

Abstract: An apparatus for processing an audio signal includes a configurable first audio signal processor for processing the audio signal in accordance with different configuration settings to obtain a processed audio signal, wherein the apparatus is adapted so that different configuration settings result in different sampling rates of the processed audio signal. The apparatus furthermore includes n analysis filter bank having a first number of analysis filter bank channels, a synthesis filter bank having a second number of synthesis filter bank channels, a second audio processor being adapted to receive and process an audio signal having a predetermined sampling rate, and a controller for controlling the first number of analysis filter bank channels or the second number of synthesis filter bank channels in accordance with a configuration setting.

Abstract: An encouraging speech system includes: at least one of reaction detection means for detecting a user's reaction in response to an encouraging speech and evaluation acquisition means for acquiring an evaluation by an evaluator on a timing and a content of the encouraging speech; good/bad determination means for determining whether the timing and the content of the encouraging speech are good or not based on at least one of the detected reaction and the acquired evaluation; learning means for learning learning data in which the timing and the content of the encouraging speech determined to be good by the good/bad determination means are associated with trigger information which serves as a trigger when this encouraging speech is made; and encouraging speech means for making the encouraging speech based on the trigger information by the user and the results of the learning by the learning means.

Abstract: Embodiments of the disclosure generally relate to a dialog system allowing for automatically reactivating a speech acquiring mode after the dialog system delivers a response to a user request. The reactivation parameters, such as a delay, depend on a number of predetermined factors and conversation scenarios. The embodiments further provide for a method of operating of the dialog system. An exemplary method comprises the steps of: activating a speech acquiring mode, receiving a first input of a user, deactivating the speech acquiring mode, obtaining a first response associated with the first input, delivering the first response to the user, determining that a conversation mode is activated, and, based on the determination, automatically re-activating the speech acquiring mode within a first predetermined time period after delivery of the first response to the user.