Nippon Electric Co., Ltd.

Seibi Chiba

Hiroaki Sakoe

Masao Watari

A pattern recognition apparatus for recognizing spoken words of a nonspecific speaker or of a specific speaker. A reference pattern composed of a sequence of feature vectors, each composed of n feature parameters, b.sub.i, is stored. The reference pattern represents a form of average of said words to be recognized as determined by multiple reference speakers speaking the same words or by the specified speaker speaking said words several times. A deviation pattern composed of a sequence of feature vectors composed of n feature parameters, w.sub.i /2, is stored. The deviation pattern is a measure of the deviation from the reference of the repeated utterances of the reference speakers or the specified speaker. An input pattern, representing the utterances of a speaker, is composed of a sequence of feature vectors, each composed of n feature parameters, a.sub.i, and is stored. A measure of the similarity of the input and reference patterns is calculated, taking into account the deviation pattern, according to one of several possible distance formulae. Basically, a distance parameter calculated for each corresponding input, reference and deviation parameter is set to zero value if the input parameter is inside the deviation range of the reference parameter, and is otherwise calculated to be a finite value.

Pattern distance calculating equipment

A system for recognizing a continuously spoken word sequence with reference to preselected reference words with the problem of coarticulation removed, comprises a pattern memory for memorizing demi-word pair reference patterns consisting of a former and a latter reference pattern segment for each reference word and a word pair reference pattern segment for each permutation with repetition of two words selected from the preselected reference words. A recognition unit is operable as a finite-state automaton on concatenating the demi-word pair reference patterns so that no contradiction occurs at each interface of the reference patterns in every concatenation. It is possible to use the automaton in restricting the number of reference patterns in each concatenation either to an odd or an even positive integer.

System operable as an automaton for recognizing continuously spoken

A multiple channel voice I/O system including a control device and switching mechanism for enabling the system to handle a greater number of user channels than there are recognition inputs. The control unit responds to two types of commands from a host system, the first type indicating a voice output operation and the second type indicating a voice input/recognition operation. During voice output for any user channel, that channel is not connected to any input terminal of the recognition unit. During voice input/recognition operation, the control unit searches for an unoccupied input terminal and connects it to a user channel.

Multiple-channel voice input/output system

In a speech recognition system of the type including a recognition unit responsive to a voice input and a conditioning input for recognizing the voice input to produce a recognition output, a start signal is produced whenever a voice input exceeds a threshold level and a pause interval detection signal is produced whenever a voice input falls below a threshold level. An output timing signal is produced when the detection signal lasts a preselected interval of time that may be either about 250 milliseconds or about 250 milliseconds plus a delay. The recognition output from the recognition unit produced in response to the detection signal is displayed in response also to the detection signal. The result is delivered to a utilization device in response to the output timing signal. The delay may be given either by a predetermined duration or an interval between those instants at which the above-mentioned 250 milliseconds have just elapsed after production of the detection signal and after production of another pause interval detection signal for a next following voice input. During the delay, it is possible either by a manually operable switch or a cancel voice input to cancel delivery of the recognition result displayed to be incorrect.

Speech recognition system with delayed output

A frequency analyzer of a predetermined number of channels comprises a digital band-pass and a digital low-pass filter section. A preselected number of digital samples sampled with a sampling period are stored in a buffer memory to be read out in each sampling period. A band-pass parameter memory memorizes for the respective channels band-pass filter impulse responses, each having preselected values, equal in number to the preselected number. A band-pass filter calculator calculates a convolution of each read-out digital sample and the preselected values of each response. The low-pass filter section produces frequency analysis results for the respective channels in each analyzing frame period equal to a first prescribed number of the sampling periods. A low-pass filter impulse response having a duration equal to the frame period multiplied by a second prescribed number and prescribed values, the predetermined number in each sampling period, is memorized in a low-pass parameter memory with its phase or the duration successively shifted by the frame period. A multiplier calculates products of each convolution and the prescribed values of the respective memorized responses read out in an interval of production of that convolution. Absolute values of the products are accumulated in an accumulator register memory according to the respective channels and the respective phases and produced therefrom as the analysis results for the respective channels in those of the sampling periods which are spaced by the frame period and at which the respectively shifted durations have ends, respectively.

Growing community of inventors

Tokyo, Japan

Seibi Chiba