Speech defects are typically addressed by having the patient or learner undergo several sessions with speech therapists, who apply specialized therapeutic tools. Speech therapies tend to be expensive, require the scheduling of appointments, and do not lend themselves easily to self-paced self-improvement.
This disclosure presents techniques that automatically provide speech-improvement feedback, thereby enabling self-managed speech therapy. Given a speech utterance by a user, the techniques cause display of a sequence of images of speech-organ positions, e.g., tongue, lips, throat muscles, etc., that correspond to the actual utterance as well as a targeted, ideal utterance. Further phonetic feedback is provided to the user using visual, tactile, spectrogram, or other modes, such that a speaker who is hard of learning can work towards a target pronunciation. The techniques also apply to foreign language learning.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Kanevsky, Dimitri; Savla, Sagar; and Starner, Thad, "Self-managed Speech Therapy", Technical Disclosure Commons, (August 14, 2018)