Abstract: In one example, the present disclosure describes a device, computer-readable medium, and method for automatically learning and facilitating interaction routines involving at least one human participant. In one example, a method includes learning an interaction routine conducted between a human user and a second party, wherein the interaction routine comprises a series of prompts and responses designed to identify and deliver desired information, storing a template of the interaction routine based on the learning, wherein the template includes at least a portion of the series of prompts and responses, detecting, in the course of a new instance of the interaction routine, at least one prompt from the second party that requests a response from the human user, and using the template to provide a response to the prompt so that involvement of the human user in the new instance of the interaction routine is minimized.

Abstract: Systems and methods for processing speech transcription in a speech processing system are disclosed. A first transcription of a first utterance is received. In response to receiving an indication of an erroneous transcribed word in the first transcription, a control circuitry automatically activates an audio receiver for receiving a second utterance. In response to receiving the second utterance, an audio file of the second utterance and an indication of a location of the erroneous transcribed word within the first transcription is transmitted to a speech recognition system for a second transcription of the second utterance. Subsequently, the erroneous transcribed word in the first transcription is replaced with a transcribed word from the second transcription.

Abstract: An on-board agent system includes: a plurality of agent functional units, each of the plurality of agent functional units being configured to provide a service including outputting a response using voice to an output unit according to an utterance of an occupant of a vehicle; and a common operator configured to be shared by the plurality of agent functional units and provided in the vehicle, wherein, when an operation is executed on the common operator with an operation pattern set to correspond to each of the plurality of agent functional units, an agent functional unit corresponding to the operation pattern of the executed operation is activated.

Abstract: Embodiments of the present disclosure provide a method and apparatus for processing a speech. The method may include: acquiring an original speech; performing speech recognition on the original speech, to obtain an original text corresponding to the original speech; associating a speech segment in the original speech with a text segment in the original text; recognizing an abnormal segment in the original speech and/or the original text; and processing a text segment indicated by the abnormal segment in the original text and/or the speech segment indicated by the abnormal segment in the original speech, to generate a final speech. A speech segment in the original speech is associated with a text segment in the original text to realize visual processing of the speech.

Abstract: A voice control method, includes: acquiring a voice input information; recognizing the voice input information to obtain a voice command; based on the voice command, determining a control corresponding to the voice command by a test framework calling unit, where the test framework calling unit is not in an application program in which the control is coded; and executing a function corresponding to the control. A voice control device and a computer-executable non-volatile storage medium are further provided.

Abstract: A method for decoding a digital signal encoded using predictive coding and transform coding, comprising the following steps: predictive decoding of a preceding frame of the digital signal, encoded by a set of predictive coding parameters; detecting the loss of a current frame of the encoded digital signal; generating by prediction, from at least one predictive coding parameter encoding the preceding frame, a frame for replacing the current frame; generating by prediction, from at least one predictive coding parameter encoding the preceding frame, an additional segment of digital signal; temporarily storing said additional segment of digital signal.

Abstract: A computing device comprising a processor, the processor configured to: receive, from an image capture system, an image captured in an environment and image metadata associated with the image, the image metadata comprising an image capture time; receive a sound recognition message from a sound recognition module, the sound recognition message comprising (i) a sound recognition identifier indicating a target sound or scene that has been recognised based on captured audio data captured in the environment, and (ii) time information associated with the sound recognition identifier; detect that the target sound or scene occurred at a time that the image was captured based on the image metadata and the time information in the sound recognition message; and output a camera control command to said image capture system based on said detection.

Abstract: An audio decoder for providing a decoded audio information includes a arithmetic decoder for providing a plurality of decoded spectral values on the basis of an arithmetically-encoded representation of the spectral values and a frequency-domain-to-time-domain converter for providing a time-domain audio representation using the decoded spectral values. The arithmetic decoder is configured to select a mapping rule describing a mapping of a code value onto a symbol code in dependence on a context state. The arithmetic decoder is configured to determine or modify the current context state in dependence on a plurality of previously-decoded spectral values. The arithmetic decoder is configured to detect a group of a plurality of previously-decoded spectral values, which fulfill, individually or taken together, a predetermined condition regarding their magnitudes, and to determine the current context state in dependence on a result of the detection. An audio encoder uses similar principles.

Abstract: An autonomously motile device may be controlled by speech received by a user device. A first speech-processing system associated with the user device may determine that audio data includes a representation of a command; a second speech-processing system associated with the autonomously motile device may determine that the command should be executed by the autonomously motile device. A network connection is established between the user device and the autonomously motile device, and a device manager authorizes execution of the command.

Abstract: According to one embodiment, a dialogue system includes a setting apparatus and a processing apparatus. The setting apparatus sets in advance a plurality of words that are in impossible combination relationships to each other. The processing apparatus acquires speech of a user, and when a speech recognition result of an object included in the speech includes a word combination included in the plurality of words that are in impossible combination relationships to each other, output a notification to the user that processing of the object cannot be carried out.

Abstract: Techniques for initiating system actions based on conversational content are disclosed. A system identifies a first conversational moment type. The first conversational moment type is defined by a first set of one or more conversational conditions. The system receives a user-selected action to be performed by the system in response to detecting conversational moments of the first conversational moment type. The system stores the user-selected action in association with the first conversational moment type. The system performs the user-selected action in response to detecting the conversational moments of the first conversational moment type.

Abstract: Audio data saved at the end of client interactions are sampled, analyzed for pauses in speech, and sliced into stretches of acoustic data containing human speech between those pauses. The acoustic data are accompanied by machine transcripts made by VoiceAI. A suitable distribution of data useful for training and testing are stipulated during data sampling by applying certain filtering criteria. The resulting datasets are sent for transcription by a human transcriber team. The human transcripts are retrieved, some post-transcription processing and cleaning are performed, and the results are added to datastores for training and testing an acoustic model.

Abstract: In an audio signal, one or more processing circuits recognize spoken content in a user's own speech signal using speech recognition and natural language understanding. The spoken content describes a listening difficulty of the user. The one or more processing circuits generate, based on the spoken content, one or more actions for hearing devices and feedback for the user. The one or more actions attempt to resolve the listening difficulty. Additionally, the one or more processing circuits convert the user feedback to verbal feedback using speech synthesis and transmit the one or more actions and the verbal feedback to the hearing devices via a body-worn device. The hearing devices are configured to perform the one or more actions and play back the verbal feedback to the user.

Abstract: This disclosure proposes systems and methods employing dynamic skill endpoints by allowing skills to register themselves with a language processing system. The language processing system allows the skill system to open a persistent network connection to the language processing system. This connection does not require the machine(s) running the skill system to have an Internet routable address; rather the skill system can contact the language processing system, which can remain at a static address, through any local routers or firewalls which may block connections from being initiated from outside the local area network. This registration opens the connection between the skill system and the language processing system. When the language processing system receives a skill invocation request indicating the skill, the language processing system can check its registry for a dynamic endpoint corresponding to the skill, and route the request over the network connection to the registered endpoint.

Abstract: An audio encoder for encoding an audio signal includes: a first encoding processor for encoding a first audio signal portion in a frequency domain, wherein the first encoding processor includes: a time frequency converter for converting the first audio signal portion into a frequency domain representation having spectral lines up to a maximum frequency of the first audio signal portion; a spectral encoder for encoding the frequency domain representation; a second encoding processor for encoding a second different audio signal portion in the time domain; a cross-processor for calculating, from the encoded spectral representation of the first audio signal portion, initialization data of the second encoding processor, so that the second encoding processing is initialized to encode the second audio signal portion immediately following the first audio signal portion in time in the audio signal; a controller configured for analyzing the audio signal and for determining, which portion of the audio signal is the firs

Abstract: An apparatus includes processor(s) to: generate a set of candidate n-grams based on probability distributions from an acoustic model for candidate graphemes of a next word most likely spoken following at least one preceding word spoken within speech audio; provide the set of candidate n-grams to multiple devices; provide, to each node device, an indication of which candidate n-grams are to be searched for within the n-gram corpus by each node device to enable searches for multiple candidate n-grams to be performed, independently and at least partially in parallel, across the node devices; receive, from each node device, an indication of a probability of occurrence of at least one candidate n-gram within the speech audio; based on the received probabilities of occurrence, identify the next word most likely spoken within the speech audio; and add the next word most likely spoken to a transcript of the speech audio.

Abstract: A communication terminal is communicable with a conversion system. The communication terminal includes circuitry configured to: receive a selection of one of a first mode and a second mode, the first mode being a mode in which audio data obtained based on sound collected by a sound collecting device is converted into text data, the second mode being a mode in which audio data obtained based on sound to be output from a sound output device is converted into text data, the audio data being relating to content obtained during an event being conducted; transmit, to the conversion system, audio data corresponding to selected one of the first mode and the second mode; receive, from the conversion system, text data converted from the transmitted audio data; and control a display to display text based on the received text data.

Abstract: A conversational system that recognizes, understands, and acts on multiple intents that may be explicit or implicit during conversations with humans. During a conversation, one or more utterances are received and processed through a plurality of machine learning algorithms to establish precise meanings, additional intentions, and alternative hypothesis. Using a combination of machine learning algorithms and datastores, conversations are interpreted as intended and may diverge where needed or desired, delivering a more useful, natural, and human-like dialogue between machines and people.

Abstract: A system and method are disclosed for setting up a communication link between a device or application and a system with a controller. The controller can collect and send information to the application. A user interfaces with the controller to access the functionality of the application through providing commands to the controller. The system allows the user to interface with multiple applications.

Abstract: A method for playing voice, which is applied to a webcast server, the method comprising: receiving voice data sent by at least one first electronic device for obtaining a voice data set, the first electronic device having a first preset authority, and the voice data set comprising at least one piece of the voice data; receiving audio-video data sent by a second electronic device, the second electronic device having a second preset authority, the audio-video data comprising the voice data selected for playback, wherein the voice data selected for playback comprises any voice data of the voice data set clicked for playback; pushing the audio-video data to each first electronic device.