System for enhancing speech performance via pattern detection and learning

Abstract

A method, system, and computer product for enhancing speech performance include includes communicating, via an input/output (I/O) device, speech data of a patient with speech problems, segmenting the speech data, generating one or more feature vectors based on at least the segmented speech data, determining whether the one or more feature vectors match with one or more recognition objects pre-trained using clinical data of one or more other patients, determining a speech disorder based on a matched result between the one or more feature vectors and the one or more recognition objects, and communicating, via the I/O device, one or more ameliorative actions for mitigating the determined speech disorder.

1 . A system for enhancing speech performance, comprising:
 an input/output (I/O) device communicating speech data of a patient;   a speech analyzer device segmenting the speech data;   a speech recognition device;   a processing device;   a memory device; and   a bus operably coupling devices,   the speech recognition device generating one or more feature vectors based on at least the segmented speech data, determining whether the one or more generated feature vectors match with one or more recognition objects pre-trained using clinical data collected from one or more other patients, and determining a speech disorder based on a matched result between the one or more feature vectors and the one or more recognition objects; and   the processing device communicating, via the I/O device, one or more ameliorative actions for mitigating the determined speech disorder.   
     
     
         2 . The system of  claim 1 , wherein the speech recognition device is based on a deep neural network (DNN). 
     
     
         3 . The system of  claim 1 , wherein the I/O device is implemented using at least one of a microphone, a headphone, a speaker, a smart watch, and an artificial intelligent (AI) listener device. 
     
     
         4 . The system of  claim 3 , wherein the AI listener device is trained depending on a degree of severity of speech order. 
     
     
         5 . The system of  claim 1 , wherein the AI listener device comprises: a stuttered region identification block identifying a stuttered region of speech data and a stuttered region reconstruction block reconstructing the stuttering region to provide a smooth speech signal. 
     
     
         6 . The system of  claim 1 , further comprising:
 a therapeutic device assisting the patient in practicing the communicated one or more ameliorative actions.   
     
     
         7 . The system of  claim 1 , further comprising:
 an N-dimensional database (N is an integer greater than 1) storing at least one of an ameliorative action, a therapeutic device, an alternative communication (AAC) device, and an identification of a healthcare professional which are appropriate to the patient,   wherein the at least one of the ameliorative action, the therapeutic device, and the identification of the healthcare professional is indexed by a combination of at least two selected from N parameters,   wherein the N parameters comprise:   a determined speech disorder for the patient;   whether the patient is alone or with a caregiver or aid;   physical characteristics of the patient;   whether the patient is familiar with the system, the therapeutic device or the AAC device;   a progression of problems for the patient;   a history of the problems for the patient;   a progression of problems for a cohort associated with the patient;   a history of the problems for the cohort; and   patient context data.   
     
     
         8 . The system of  claim 1 , wherein the processing device communicating, via the I/O device, one or more ameliorative actions for mitigating the determined speech disorder further comprises using data selected from a group consisting of: patient context data, the clinical data, patient physical and emotional condition data, and patient progress data. 
     
     
         9 . The system of  claim 8 , wherein the patient context data is selected from a group consisting of:
 a family background, language environment, age, gender, occupation, culture, and residential region.   
     
     
         10 . The system of  claim 7 , wherein a list of the parameters or an output indexed by the combination of at least two parameters selected from the parameters in the N-dimensional database is changed based on the system learning what is more effective for the patient or the cohort associated with the patient. 
     
     
         11 . The system of  claim 1 , wherein the speech recognition device generates the one or more feature vectors based on patient context data. 
     
     
         12 . The system of  claim 1 , wherein the I/O device receives the speech data from the patient responsive to one or more communications via the I/O device, the one or more communications generated based on patient context data. 
     
     
         13 .- 16 . (canceled) 
     
     
         17 . A computer program product comprising a computer-readable storage medium having computer readable program instructions embodied therewith, the computer readable program instructions executable by at least one processor to cause a computer to perform a computer-implemented method comprising:
 communicating, via an input/output (I/O) device, speech data of a patient;   segmenting the speech data;   generating one or more feature vectors based on at least the segmented speech data;   determining whether the one or more feature vectors match with one or more recognition objects pre-trained using clinical data collected from one or more other patients;   determining a speech disorder based on a matched result between the one or more feature vectors and the one or more recognition objects; and   communicating, via the I/O device, one or more ameliorative actions for mitigating the determined speech disorder.   
     
     
         18 . The computer program product of  claim 17 , wherein the speech recognition device is implemented based on a deep neural network (DNN). 
     
     
         19 . The computer program product of  claim 17 , wherein the communicating, via the I/O device, the one or more ameliorative actions further comprises using data selected from a group consisting of: the patient context data, the clinical data, patient physical and emotional condition data, and patient progress data. 
     
     
         20 . The computer program product of  claim 17 , wherein the one or more feature vectors are generated further based on patient context data.