Abstract: A speech detection system compares the amplitude of an audio signal during successive time periods with speech detection thresholds, and generates an indication of whether the signal contains speech. It derives a background amplitude level from portions of the signal which it indicates do not contain speech, and improves its speech detection by altering the amplitude of the audio signal relative to the speech detection thresholds as a function of this background level. Preferably the background amplitude level is a moving average, which is repeatedly recalculated and repeatedly used to alter the relative amplitude of the audio signal and the detection thresholds. The apparatus uses a measure of the variability of the background amplitude to improve its speech detection. It generates start-of-speech and end-of-speech indications when the amplitude crosses respective thresholds for specified numbers of frames.

Abstract: A text locating system recognizes spoken utterances, uses the recognized words as a search string, and searches text for words matching that search string. The probability that a given vocabulary word is selected as a search word is altered both by limiting the recognizable vocabulary to words in the text to the searched, and by altering the probability that individual recognizable words will be selected as a function of the number of time they occur in that text. The system performs incremental searches by adding successively recognized words to the search string and searching for the next occurrence of the string in response to each such addition. The invention can be used in a text editing system which enables a user to switch between a dictation mode, which inserts recognized words into text, and a search mode, which uses them to search for new cursor locations.

Abstract: A method determines if a portion of speech corresponds to a speech pattern by time aligning both the speech and a plurality of speech pattern models against a common time-aligning model. This compensates for speech variation between the speech and the pattern models. The method then compares the resulting time-aligned speech model against the resulting time-aligned pattern models to determine which of the patterns most probably corresponds to the speech. Preferably there are a plurality of time-aligning models, each representing a group of somewhat similar sound sequences which occur in different words. Each of these time-aligning models is scored for similarity against a portion of speech, and the time-aligned speech model and time-aligned pattern models produced by time alignment with the best scoring time-aligning model are compared to determine the likelihood that each speech pattern corresponds to the portion of speech. This is performed for each successive portion of speech.

Abstract: A method of speech analysis calculates one or more difference parameters for each of a sequence of acoustic frames, where each difference parameter is a function of the difference between an acoustic parameter in one frame and an acoustic parameter in a nearby frame. The method is used in speech recognition which compares the difference parameters of each frame against acoustic models representing speech units, where each speech-unit model has a model of the difference parameters associated with the frames of its speech unit. The difference parameters can be slope parameters or energy difference parameters. Slope parameters are derived by finding the difference between the energy of a given spectral parameter of a given frame and the energy, in a nearby frame, of a spectral parameter associated with a different frequency band. The resulting parameter indicates the extent to which the frequency of energy in the part of the spectrum represented by the given parameter is going up or going down.

Abstract: Smoothed frame labeling associates phonetic frame labels with a given speech frame as a function of (a) the closeness with which the given frame compares to each of a plurality of acoustic models, (b) which frame labels correspond with a neighboring frame, and (c) transition probabilities which indicate, for the frame labels associated with the neighboring frame, which frame labels are probably associated with the given frame. The smoothed frame labeling is used to divide the speech into segments of frames having the same class of labels. The invention represents words as a collection of known diphone models, each of which models the sound before and after a boundary between segments derived by the smoothed frame labeling.

Abstract: A system is disclosed for recognizing a pattern in a collection of data given a context of one or more other patterns previously identified. Preferably the system is a speech recognition system, the patterns are words and the collection of data is a sequence of acoustic frames. During the processing of each of a plurality of frames, for each word in an active vocabulary, the system updates a likelihood score representing a probability of a match between the word and the frame, combines a language model score based on one or more previously recognized words with that likelihood score, and prunes the word from the active vocabulary if the combined score is below a threshold. A rapid match is made between the frames and each word of an initial vocabulary to determine which words should originally be placed in the active vocabulary.