Automatic speech recognition

A method of speech recognition, voice recognition, and a combination thereof, sequentially executed by at least one processor, on a plurality of consecutive segments, the method comprising:

• obtaining digital information of a segment selected from the group consisting of speech segment, voice segment, and a combination thereof, wherein the digital information comprises a spectrogram representation;
• extracting a plurality of features characterizing the segment from the spectrogram representation;
• determining a consistent structure segment vector based on the features;
• deploying machine learning to determine at least one label of the segment vector;
• and outputting the at least one label;
• wherein the extracting a plurality of features further comprises assembling a plurality of matrixes and an index matrix, having identical number of cells, wherein each matrix of the plurality of matrixes represent a different feature of the plurality of features, wherein assembling the index matrix is based on a spectrogram having time frames and frequency bands, wherein the index matrix dimensions correlates with the time frames and frequency bands of the spectrogram, wherein the plurality of matrixes overlap with the index matrix, and wherein a content of each cell of each matrix of the plurality of matrixes represents a feature value of a time frame and a frequency band indicated by the index matrix, wherein contiguous time frames containing similar features values are replaced with a time interval in the index matrix and the plurality of matrixes.

A method of speech recognition, voice recognition, and a combination thereof, sequentially executed by at least one processor, on a plurality of consecutive segments, the method comprising:

• obtaining digital information of a segment selected from the group consisting of speech segment, voice segment, and a combination thereof, wherein the digital information comprises a spectrogram representation;
• extracting a plurality of features characterizing the segment from the spectrogram representation;
• determining a consistent structure segment vector based on the features;
• deploying machine learning to determine at least one label of the segment vector;
• and outputting the at least one label;
• wherein the extracting a plurality of features further comprises assembling a plurality of matrixes and an index matrix, having identical number of cells, wherein each matrix of the plurality of matrixes represent a different feature of the plurality of features, wherein assembling the index matrix is based on a spectrogram having time frames and frequency bands, wherein the index matrix dimensions correlates with the time frames and frequency bands of the spectrogram, wherein the plurality of matrixes overlap with the index matrix, and wherein a content of each cell of each matrix of the plurality of matrixes represents a feature value of a time frame and a frequency band indicated by the index matrix, wherein the determining a consistent structure segment vector further comprises compiling a plurality of components each comprising equal number of operands, wherein the first component of the plurality of components is an index component corresponding with the index matrix while the rest of the plurality of components are features components corresponding with the features matrixes, wherein a total number of operands is all possible combinations of frequency bands pairs, and wherein the index component indicates operands having band pairs presence in the segment vector.