Abstract: A system, article, and method provide temporal-domain feature extraction for automatic speech recognition.

Abstract: Techniques related to a method and system of robust speaker recognition activation are described herein. Such techniques apply keyphrase detection and speaker recognition to a subsequent phrase after detecting a waking keyphrase.

Abstract: An automatic speech recognition (ASR) system includes a memory configured to store a filler model. The filler model includes one or more phonetic strings corresponding to one or more portions of a wake up phrase. The ASR system also includes one or more processors operatively coupled to the memory and configured to analyze a speech signal with the filler model to determine whether the speech signal includes the wake up phrase or any portion of the wake up phrase. The one or more processors are also configured to generate, based on the analysis, a hypothesis of underlying speech included in the speech signal. The hypothesis excludes the wake up phrase or any portion of the wake up phrase included in the speech signal.

Abstract: Techniques are provided for acoustic event detection. A methodology implementing the techniques according to an embodiment includes extracting acoustic features from a received audio signal. The acoustic features may include, for example, one or more short-term Fourier transform frames, or other spectral energy characteristics, of the audio signal. The method also includes applying a trained classifier to the extracted acoustic features to identify and label acoustic event subparts of the audio signal and to generate scores associated with the subparts. The method further includes performing sequence decoding of the acoustic event subparts and associated scores to detect target acoustic events of interest based on the scores and temporal ordering sequence of the event subparts. The classifier is trained on acoustic event subparts that are generated through unsupervised subspace clustering techniques applied to training data that includes target acoustic events.

Abstract: Techniques are provided for segmentation of a key phrase. A methodology implementing the techniques according to an embodiment includes accumulating feature vectors extracted from time segments of an audio signal, and generating a set of acoustic scores based on those feature vectors. Each of the acoustic scores in the set represents a probability for a phonetic class associated with the time segments. The method further includes generating a progression of scored model state sequences, each of the scored model state sequences based on detection of phonetic units associated with a corresponding one of the sets of acoustic scores generated from the time segments of the audio signal. The method further includes analyzing the progression of scored state sequences to detect a pattern associated with the progression, and determining a starting and ending point for segmentation of the key phrase based on alignment of the detected pattern with an expected pattern.

Abstract: Speech or non-speech detection techniques are discussed and include updating a speech pattern model using probability scores from an acoustic model to generate a score for each state of the speech pattern model, such that the speech pattern model includes a first non-speech state having multiple self loops each associated with a non-speech probability score of the probability scores, a plurality of speech states following the first non-speech state, and a second non-speech state following the speech states, and detecting speech based on a comparison of a score of the first non-speech state and a score of the last speech state of the multiple speech states.

Abstract: Techniques are provided for wake-on-voice (WOV) key-phrase enrollment. A methodology implementing the techniques according to an embodiment includes generating a WOV key-phrase model based on identification of the sequence of sub-phonetic units of a user-provided key-phrase. The WOV key-phrase model is employed by a WOV processor for detection of the user spoken key-phrase and triggering operation of an automatic speech recognition (ASR) processor in response to the detection. The method further includes updating an ASR language model based on the user-provided key-phrase. The update includes one of embedding the WOV key-phrase model into the ASR language model, converting sub-phonetic units of the WOV key-phrase model and embedding the converted WOV key-phrase model into the ASR language model, or generating an ASR key-phrase model by applying a phoneme-syllable based statistical language model to the user-provided key-phrase and embedding the generated ASR key-phrase model into the ASR language model.

Abstract: Techniques are provided for efficient acoustic event detection with reduced resource consumption. A methodology implementing the techniques according to an embodiment includes calculating frames of power spectra based on segments of received acoustic signals. The method further includes two processes, one for detecting impulsive acoustic events and another for detecting continuous acoustic events. The first process includes generating impulsive acoustic event features associated with first and second power spectrum frames, applying a neural network classifier to the impulsive acoustic event features to generate event scores, and detecting an impulsive acoustic event based on those event scores.

Abstract: Key phrase detection techniques for applications such as wake on voice are discussed include performing a vectorized operation on a multiple element acoustic score vector for a current time instance including a single state rejection model score and scores for a multiple state key phrase model and a multiple element state score vector for a previous time instance including a previous state score for the single state rejection model and previous state scores for the multiple state key phrase model to generate a multiple element score summation vector and a second vectorized operation on the multiple element score summation vector to determine a multiple element state score vector for the current time instance. The multiple element state score vector for the current time instance may then be evaluated to determine whether received audio input includes a key phrase corresponding to the multiple state key phrase model.

Abstract: Techniques are provided for language identification performed in conjunction with wake-on-voice keyword detection. A methodology implementing the techniques according to an embodiment includes applying phrase models to a user-spoken keyword. Each of the phrase models is configured to detect the keyword in a selected language and to generate a probability associated with the detection. The method further includes scoring the probabilities associated with the keyword detection in each of the languages, and identifying the language of the keyword based on the scoring. Automatic speech recognition and spoken language understanding systems may then be configured or selected to process further speech from the user in the identified language. In some embodiments, the phrase models are generated, in an offline process, based on provided grapheme sequences representing the keyword in the language associated with the phrase model.

Abstract: Key phrase detection techniques for applications such as wake on voice are discussed include performing a vectorized operation on a multiple element acoustic score vector for a current time instance including a single state rejection model score and scores for a multiple state key phrase model and a multiple element state score vector for a previous time instance including a previous state score for the single state rejection model and previous state scores for the multiple state key phrase model to generate a multiple element score summation vector and a second vectorized operation on the multiple element score summation vector to determine a multiple element state score vector for the current time instance. The multiple element state score vector for the current time instance may then be evaluated to determine whether received audio input includes a key phrase corresponding to the multiple state key phrase model.

Abstract: Cepstral variance normalization is described for audio feature extraction.

Abstract: Techniques are provided for spoken language understanding based on keyword spotting and speech recognition. A methodology implementing the techniques according to an embodiment includes detecting a user spoken keyword or key-phrase embedded in an initial segment of a received audio signal, which is stored in a buffer. The method further includes triggering an automatic speech recognition (ASR) processor in response to the key-phrase detection. The method further includes performing automatic speech recognition, by the ASR processor, on a combination of the buffered initial segment and one or more additional received segments of the audio signal which include further speech from the user. The method still further includes performing natural language understanding on the recognized speech to determine a user request. The key-phrase is user selectable and serves to wake the ASR processor from a sleeping or idle lower power consumption state, into an active higher power consumption recognition state.

Abstract: Techniques related to speaker recognition are discussed. Such techniques include determining context aware confidence values formed of false accept and false reject rates determined by using adaptively updated acoustic environment score distributions matched to current score distributions.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for distributed automatic speech recognition. An example apparatus includes a detector to process an input audio signal and identify a portion of the input audio signal including a sound to be evaluated, the sound to be evaluated organized into a plurality of audio features representing the sound. The example apparatus includes a quantizer to process the audio features using a quantization process to reduce the audio features to generate a reduced set of audio features for transmission. The example apparatus includes a transmitter to transmit the reduced set of audio features over a low-energy communication channel for processing.

Abstract: Key phrase detection techniques for applications such as wake on voice are discussed include performing a vectorized operation on a multiple element acoustic score vector for a current time instance including a single state rejection model score and scores for a multiple state key phrase model and a multiple element state score vector for a previous time instance including a previous state score for the single state rejection model and previous state scores for the multiple state key phrase model to generate a multiple element score summation vector and a second vectorized operation on the multiple element score summation vector to determine a multiple element state score vector for the current time instance. The multiple element state score vector for the current time instance may then be evaluated to determine whether received audio input includes a key phrase corresponding to the multiple state key phrase model.

Abstract: Techniques related to key phrase detection for applications such as wake on voice are discussed. Such techniques may include updating a start state based rejection model and a key phrase model based on scores of sub-phonetic units from an acoustic model to generate a rejection likelihood score and a key phrase likelihood score and determining whether received audio input is associated with a predetermined key phrase based on the rejection likelihood score and the key phrase likelihood score.

Abstract: Techniques related to key phrase detection for applications such as wake on voice are discussed. Such techniques may include determining a sequence of audio units for received audio input representing a user defined key phrase, eliminating audio units from the sequence to generate a final sequence of audio units, and generating a key phrase recognition model representing the user defined key phrase based on the final sequence.

Abstract: Key phrase detection techniques for applications such as wake on voice are discussed include performing a vectorized operation on a multiple element acoustic score vector for a current time instance including a single state rejection model score and scores for a multiple state key phrase model and a multiple element state score vector for a previous time instance including a previous state score for the single state rejection model and previous state scores for the multiple state key phrase model to generate a multiple element score summation vector and a second vectorized operation on the multiple element score summation vector to determine a multiple element state score vector for the current time instance. The multiple element state score vector for the current time instance may then be evaluated to determine whether received audio input includes a key phrase corresponding to the multiple state key phrase model.

Abstract: Techniques related to key phrase detection for applications such as wake on voice are discussed. Such techniques may include intermediate scoring of a state or states of a key phrase model and/or a backward transition or rejection loopback from a state of the key phrase model to a rejection model to reduce false accepts based on received utterances.