Abstract: Systems and methods are provided for speech recognition. For example, audio characteristics are extracted from acquired voice signals; a syllable confusion network is identified based on at least information associated with the audio characteristics; a word lattice is generated based on at least information associated with the syllable confusion network and a predetermined phonetic dictionary; and an optimal character sequence is calculated in the word lattice as a speech recognition result.

Abstract: A method is performed at a device having one or more processors and memory. The device establishes a first-level Deep Neural Network (DNN) model based on unlabeled speech data, the unlabeled speech data containing no speaker labels and the first-level DNN model specifying a plurality of basic voiceprint features for the unlabeled speech data. The device establishes a second-level DNN model by tuning the first-level DNN model based on labeled speech data, the labeled speech data containing speech samples with respective speaker labels, wherein the second-level DNN model specifies a plurality of high-level voiceprint features. Using the second-level DNN model, registers a first high-level voiceprint feature sequence for a user based on a registration speech sample received from the user. The device performs speaker verification for the user based on the first high-level voiceprint feature sequence registered for the user.

Abstract: A method and a device for training a DNN model includes: at a device including one or more processors and memory: establishing an initial DNN model; dividing a training data corpus into a plurality of disjoint data subsets; for each of the plurality of disjoint data subsets, providing the data subset to a respective training processing unit of a plurality of training processing units operating in parallel, wherein the respective training processing unit applies a Stochastic Gradient Descent (SGD) process to update the initial DNN model to generate a respective DNN sub-model based on the data subset; and merging the respective DNN sub-models generated by the plurality of training processing units to obtain an intermediate DNN model, wherein the intermediate DNN model is established as either the initial DNN model for a next training iteration or a final DNN model in accordance with a preset convergence condition.

Abstract: A method and device for voiceprint recognition, include: establishing a first-level Deep Neural Network (DNN) model based on unlabeled speech data, the unlabeled speech data containing no speaker labels and the first-level DNN model specifying a plurality of basic voiceprint features for the unlabeled speech data; obtaining a plurality of high-level voiceprint features by tuning the first-level DNN model based on labeled speech data, the labeled speech data containing speech samples with respective speaker labels, and the tuning producing a second-level DNN model specifying the plurality of high-level voiceprint features; based on the second-level DNN model, registering a respective high-level voiceprint feature sequence for a user based on a registration speech sample received from the user; and performing speaker verification for the user based on the respective high-level voiceprint feature sequence registered for the user.

Abstract: A method of recognizing speech is provided that includes generating a decoding network that includes a primary sub-network and a classification sub-network. The primary sub-network includes a classification node corresponding to the classification sub-network. The classification sub-network corresponds to a group of uncommon words. A speech input is received and decoded by instantiating a token in the primary sub-network and passing the token through the primary network. When the token reaches the classification node, the method includes transferring the token to the classification sub-network and passing the token through the classification sub-network. When the token reaches an accept node of the classification sub-network, the method includes returning a result of the token passing through the classification sub-network to the primary sub-network. The result includes one or more words in the group of uncommon words. A string corresponding to the speech input is output that includes the one or more words.

Abstract: An electronic device with one or more processors and memory trains an acoustic model with an international phonetic alphabet (IPA) phoneme mapping collection and audio samples in different languages, where the acoustic model includes: a foreground model; and a background model. The device generates a phone decoder based on the trained acoustic model. The device collects keyword audio samples, decodes the keyword audio samples with the phone decoder to generate phoneme sequence candidates, and selects a keyword phoneme sequence from the phoneme sequence candidates. After obtaining the keyword phoneme sequence, the device detects one or more keywords in an input audio signal with the trained acoustic model, including: matching phonemic keyword portions of the input audio signal with phonemes in the keyword phoneme sequence with the foreground model; and filtering out phonemic non-keyword portions of the input audio signal with the background model.

Abstract: A method, system and computer storage medium for visual searching based on cloud service is disclosed. The method includes: receiving, from a client, an image recognition request of cloud service, the request containing image data; forwarding, according to a set classified forwarding rule, the image data to a corresponding classified visual search service; recognizing, by the respective corresponding classified visual search services, corresponding classified type information in the image data, and determining a corresponding name of the image data in accordance with the respective classified type information, and obtaining a classified visual search result; summarizing and sending, to a client, the classified visual search result of the corresponding classified visual search service.

Abstract: A method and a device for training an acoustic language model, include: conducting word segmentation for training samples in a training corpus using an initial language model containing no word class labels, to obtain initial word segmentation data containing no word class labels; performing word class replacement for the initial word segmentation data containing no word class labels, to obtain first word segmentation data containing word class labels; using the first word segmentation data containing word class labels to train a first language model containing word class labels; using the first language model containing word class labels to conduct word segmentation for the training samples in the training corpus, to obtain second word segmentation data containing word class labels; and in accordance with the second word segmentation data meeting one or more predetermined criteria, using the second word segmentation data containing word class labels to train the acoustic language model.

Abstract: A method includes: acquiring data samples; performing categorized sentence mining in the acquired data samples to obtain categorized training samples for multiple categories; building a text classifier based on the categorized training samples; classifying the data samples using the text classifier to obtain a class vocabulary and a corpus for each category; mining the corpus for each category according to the class vocabulary for the category to obtain a respective set of high-frequency language templates; training on the templates for each category to obtain a template-based language model for the category; training on the corpus for each category to obtain a class-based language model for the category; training on the class vocabulary for each category to obtain a lexicon-based language model for the category; building a speech decoder according to an acoustic model, the class-based language model and the lexicon-based language model for any given field, and the data samples.

Abstract: The present invention provides a phase-change storage unit for replacing DRAM and FLASH and a manufacturing method thereof, and the phase-change storage unit includes a phase-change material layer and a cylindrical lower electrode being in contact with and located below the phase-change material layer, where the phase-change material layer is formed by connecting a side wall layer and a round bottom layer, forms a hollow cylinder or hollow inverted conical frustum having an opening at an upper part, and the hollow cylinder or hollow inverted conical frustum is internally filled with a medium layer.

Abstract: A method and an apparatus are provided for retrieving keyword. The apparatus configures at least two types of language models in a model file, where each type of language model includes a recognition model and a corresponding decoding model; the apparatus extracts a speech feature from the to-be-processed speech data; performs language matching on the extracted speech feature by using recognition models in the model file one by one, and determines a recognition model based on a language matching rate; and determines a decoding model corresponding to the recognition model; decoding the extracted speech feature by using the determined decoding model, and obtains a word recognition result after the decoding; and matches a keyword in a keyword dictionary and the word recognition result, and outputs a matched keyword.

Abstract: A method is implemented at a computer to determine that certain information content is composed or compiled in a specific language selected among two or more similar languages. The computer integrates a first vocabulary list of a first language and a second vocabulary list of a second language into a comprehensive vocabulary list. The integrating includes analyzing the first vocabulary list in view of the second vocabulary list to identify a first vocabulary sub-list that is used in the first language, but not in the second language. The computer then identifies, in the information content, a plurality of expressions that are included in the comprehensive vocabulary list, and a subset of expressions that are included in the first vocabulary sub-list. Upon a determination that a total frequency of occurrence of the subset of expressions meets predetermined occurrence criteria, the computer determines that the information content is composed in the first language.

Abstract: An Sb—Te—Ti phase-change thin-film material applicable to a phase-change memory and preparation thereof. The Sb—Te—Ti phase-change memory material is formed by doping an Sb—Te phase-change material with Ti, Ti forms bonds with both Sb and Te, and the Sb—Te—Ti phase-change memory material has a chemical formula SbxTeyTi100-x-y, where 0<x<80 and 0<y<100?x. When the Sb—Te—Ti phase-change memory material is a Ti—Sb2Te3 phase-change memory material, Ti atoms replace Sb atoms, and phase separation does not occur. The crystallization temperature of the Sb—Te—Ti phase-change memory material is significantly risen, retention is improved, and thermal stability is enhanced; meanwhile, the amorphous state resistance decreases, and the crystalline state resistance increases; and the Sb—Te—Ti phase-change memory material has wide application in phase-change memories.

Abstract: The present application discloses a method, an electronic system and a non-transitory computer readable storage medium for recognizing audio commands in an electronic device. The electronic device obtains audio data based on an audio signal provided by a user and extracts characteristic audio fingerprint features from the audio data. The electronic device further determines whether the corresponding audio signal is generated by an authorized user by comparing the characteristic audio fingerprint features with an audio fingerprint model for the authorized user and with a universal background model that represents user-independent audio fingerprint features, respectively. When the corresponding audio signal is generated by the authorized user of the electronic device, an audio command is extracted from the audio data, and an operation is performed according to the audio command.

Abstract: The present invention provides a phase-change storage unit containing a TiSiN material layer and a method for preparing the same. The phase-change storage unit includes a phase-change material layer and a lower electrode located there below, the phase-change material layer and the lower electrode are connected by a TiSiN material layer, the lower electrode includes a bottom and a sheet side connected to the bottom, the sheet side is perpendicular to the bottom to form a blade structure, and the top of the sheet side contacts the TiSiN material layer. The present invention adopts annealing to increase the grain size of the electrode so as to reduce the overall resistance of the device and form a TiSiN material layer on the top of the lower electrode so as to reduce the effective operation region. The phase-change storage unit of the present invention is applied to a phase-change memory to achieve the advantages such as low power consumption, high density and high data retention performance.

Abstract: A method and an apparatus are provided for retrieving keyword. The apparatus configures at least two types of language models in a model file, where each type of language model includes a recognition model and a corresponding decoding model; the apparatus extracts a speech feature from the to-be-processed speech data; performs language matching on the extracted speech feature by using recognition models in the model file one by one, and determines a recognition model based on a language matching rate; and determines a decoding model corresponding to the recognition model; decoding the extracted speech feature by using the determined decoding model, and obtains a word recognition result after the decoding; and matches a keyword in a keyword dictionary and the word recognition result, and outputs a matched keyword.

Abstract: This application discloses a method implemented of recognizing a keyword in a speech that includes a sequence of audio frames further including a current frame and a subsequent frame. A candidate keyword is determined for the current frame using a decoding network that includes keywords and filler words of multiple languages, and used to determine a confidence score for the audio frame sequence. A word option is also determined for the subsequent frame based on the decoding network, and when the candidate keyword and the word option are associated with two distinct types of languages, the confidence score of the audio frame sequence is updated at least based on a penalty factor associated with the two distinct types of languages. The audio frame sequence is then determined to include both the candidate keyword and the word option by evaluating the updated confidence score according to a keyword determination criterion.

Abstract: The present disclosure provides a method and system for realizing interaction in augmented reality. The method includes: collecting a frame image and uploads the frame image; recognizing a template image that matches the frame image and returning the template image; detecting a marker area of the frame image according to the template image; and superposing media data corresponding to the template image on the marker area and displaying the superposed image.

Abstract: A computer-implemented method is performed at a server having one or more processors and memory storing programs executed by the one or more processors for authenticating a user from video and audio data. The method includes: receiving a login request from a mobile device, the login request including video data and audio data; extracting a group of facial features from the video data; extracting a group of audio features from the audio data and recognizing a sequence of words in the audio data; identifying a first user account whose respective facial features match the group of facial features and a second user account whose respective audio features match the group of audio features. If the first user account is the same as the second user account, retrieve the sequence of words associated with the user account and compare the sequences of words for authentication purpose.

Abstract: The present invention relates to a method, device and storage medium for locating feature points on a human face. The method for locating feature points on a human face of the present invention comprises: preliminary locating of the position of a human face by combining human face detection with human eye matching, and acquiring preliminary locating information; according to the preliminary locating information, fitting feature points on the human face; and according to the fitting result, completing the locating of the feature points on the human face. The present invention has the beneficial effects that: by using human face detection and human eye matching at the same time, and incorporating at least one feature of an appearance model according to the preliminary locating information, the position information about a human face is more accurately located, and the location of the feature points on the human face is accurately accomplished.