Speech recognition system and unit

Abstract

The invention provides a speech recognition system and unit. The speech recognition system includes an image capturing device capable of capturing images of lips, an oral cavity parameter capturing device capable sensing the change of the tongue position, and a speech recognition device. The speech recognition device can receive and analyze lip shape signals of the image capturing device and tongue distance signals and jaw position signals of the oral cavity parameter capturing device to obtain a voice parameter set. An output unit outputs the voice parameter set. During a speech process of a patient with dysarthria, the speech recognition system can automatically analyze the change of the lip shapes and the tongue of the patient with dysarthria to obtain a voice parameter set corresponding to specific voice materials, and can further perform voice synthesis output according to the voice materials matching the voice parameter set to make the speeches of the patient with dysarthria clearly understood.

Description

technical field [0001] The present invention relates to a discourse recognition system and unit, in particular to a discourse recognition system for recognizing discourses spoken by patients with dysphagia, and a system for programmatically constructing electronic devices to identify patients with dysphagia The discourse recognition unit of the spoken utterance. Background technique [0002] When the average person speaks, multiple systems in the body need to coordinate with each other to speak correctly. The airflow for speaking is mainly supplied by the respiratory system. When the airflow rises through the larynx, the laryngeal system will be driven by the airflow to produce a laryngeal sound. The formed articulation system transforms laryngeal sounds into speech with different timbres and intonations, and then presents the content that the brain wants to express through speaking actions. However, when the brain, central or peripheral nerves are damaged and the function...

AI Technical Summary

Problems solved by technology

[0003] Dysphasia is not a language abnormality like aphasia, nor is it the cognitive impairment of dementia patients. Patients with dysphasia still have a correct understanding of language and know how to communicate with others through language. Because some tissues and organs of the speech system cannot function normally, the pronunciation of the spoken words is incorrect and difficult to understand

Although patients with narcolepsy cannot make clear and clear speech, every time they speak the same word, the tongue, teeth and lips in the articulation system of their mouth will still make a specific mode of operation, so each time They will all make very similar voices, and therefore, over time, people around each patient with dysphagia can still understand the correct words they want to express, but for those who rarely interact with the patient with dysphagia However, it will be quite troublesome, and embarrassing scenes of misunderstanding often occur

Images