Machine-generated speech transcriptions are helpful for people that are hard of hearing or individuals who don’t understand the spoken language to understand conversations that take place near them. Augmented reality glasses can display speech transcriptions to help users understand such conversations. However, the display of transcriptions can be distracting when the user wearing the AR glasses is herself speaking. This disclosure describes techniques to automatically identify speakers in a conversation and only display live transcriptions of speech from conversation participants other than the person to whom the transcription is being provided. Speakers may be identified using user-permitted factors, e.g., by use of a head-related transfer function (HRTF), biometric voice recognition, or facial feature (e.g., lip movement) recognition.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
N/A, "Identifying Speakers and Limiting Displayed Transcription to Select Speakers", Technical Disclosure Commons, (December 14, 2020)