Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing speaker verification. In one aspect, a method includes accessing a neural network having an input layer that provides inputs to a first hidden layer whose nodes are respectively connected to only a proper subset of the inputs from the input layer. Speech data that corresponds to a particular utterance may be provided as input to the input layer of the neural network. A representation of activations that occur in response to the speech data at a particular layer of the neural network that was configured as a hidden layer during training of the neural network may be generated. A determination of whether the particular utterance was likely spoken by a particular speaker may be made based at least on the generated representation. An indication of whether the particular utterance was likely spoken by the particular speaker may be provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for keyword spotting. One of the methods includes training, by a keyword detection system, a convolutional neural network for keyword detection by providing a two-dimensional set of input values to the convolutional neural network, the input values including a first dimension in time and a second dimension in frequency, and performing convolutional multiplication on the two-dimensional set of input values for a filter using a frequency stride greater than one to generate a feature map.

Abstract: Embodiments pertain to automatic speech recognition in mobile devices to establish the presence of a keyword. An audio waveform is received at a mobile device. Front-end feature extraction is performed on the audio waveform, followed by acoustic modeling, high level feature extraction, and output classification to detect the keyword. Acoustic modeling may use a neural network or a vector quantization dictionary and high level feature extraction may use pooling.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recognizing keywords using a long short term memory neural network. One of the methods includes receiving, by a device for each of multiple variable length enrollment audio signals, a respective plurality of enrollment feature vectors that represent features of the respective variable length enrollment audio signal, processing each of the plurality of enrollment feature vectors using a long short term memory (LSTM) neural network to generate a respective enrollment LSTM output vector for each enrollment feature vector, and generating, for the respective variable length enrollment audio signal, a template fixed length representation for use in determining whether another audio signal encodes another spoken utterance of the enrollment phrase by combining at most a quantity k of the enrollment LSTM output vectors for the enrollment audio signal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining hotword suitability. In one aspect, a method includes receiving speech data that encodes a candidate hotword spoken by a user, evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, generating a hotword suitability score for the candidate hotword based on evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, and providing a representation of the hotword suitability score for display to the user.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a deep neural network. One of the methods for training a deep neural network that includes a low rank hidden input layer and an adjoining hidden layer, the low rank hidden input layer including a first matrix A and a second matrix B with dimensions i×m and m×o, respectively, to identify a keyword includes receiving a feature vector including i values that represent features of an audio signal encoding an utterance, determining, using the low rank hidden input layer, an output vector including o values using the feature vector, determining, using the adjoining hidden layer, another vector using the output vector, determining a confidence score that indicates whether the utterance includes the keyword using the other vector, and adjusting weights for the low rank hidden input layer using the confidence score.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for key phrase detection. One of the methods includes receiving a plurality of audio frame vectors that each model an audio waveform during a different period of time, generating an output feature vector for each of the audio frame vectors, wherein each output feature vector includes a set of scores that characterize an acoustic match between the corresponding audio frame vector and a set of expected event vectors, each of the expected event vectors corresponding to one of the scores and defining acoustic properties of at least a portion of a keyword, and providing each of the output feature vectors to a posterior handling module.

Abstract: Embodiments pertain to automatic speech recognition in mobile devices to establish the presence of a keyword. An audio waveform is received at a mobile device. Front-end feature extraction is performed on the audio waveform, followed by acoustic modeling, high level feature extraction, and output classification to detect the keyword. Acoustic modeling may use a neural network or Gaussian mixture modeling, and high level feature extraction may be done by aligning the results of the acoustic modeling with expected event vectors that correspond to a keyword.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a deep neural network. One of the methods includes training a deep neural network with a first training set by adjusting values for each of a plurality of weights included in the neural network, and training the deep neural network to determine a probability that data received by the deep neural network has features similar to key features of one or more keywords or key phrases, the training comprising providing the deep neural network with a second training set and adjusting the values for a first subset of the plurality of weights, wherein the second training set includes data representing the key features of the one or more keywords or key phrases.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for key phrase detection. One of the methods includes receiving a plurality of audio frame vectors that each model an audio waveform during a different period of time, generating an output feature vector for each of the audio frame vectors, wherein each output feature vector includes a set of scores that characterize an acoustic match between the corresponding audio frame vector and a set of expected event vectors, each of the expected event vectors corresponding to one of the scores and defining acoustic properties of at least a portion of a keyword, and providing each of the output feature vectors to a posterior handling module.

Abstract: In general, the subject matter described in this specification can be embodied in methods, systems, and program products for performing compounded text segmentation. Compounded text that is extracted from one or more search queries submitted to a search engine is received. The compounded text includes a plurality of individual words that are joined together without intervening spaces. An electronic dictionary including words is accessed. A data structure representing possible segmentations of the compounded text is generated based on whether words in the possible segmentations occur in the electronic dictionary. A data store comprising data associated with a same field of usage as the compounded text is accessed to determine a frequency of occurrence for possible segmentations of the data structure. A segmentation of the compounded text that is most probable based on the data is determined. A language model is trained using the determined segmentation of the compounded text.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining hotword suitability. In one aspect, a method includes receiving speech data that encodes a candidate hotword spoken by a user, evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, generating a hotword suitability score for the candidate hotword based on evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, and providing a representation of the hotword suitability score for display to the user.