Voice and Touch Based Error-tolerant Multimodal Text Editing and Correction for Smartphones
summary
A smartphone system called VT combines touch plus speech so users tap approximate location then speak insertions or editing commands; it generates substitution and insertion candidates and scores them using a tri-gram language model, word-embedding similarity, edit distance, and phrase location to pick the top replacement. VT supports word, sentence and paragraph edits and single or compound commands, automatically inferring location without precise selection. User studies show ~30.8% faster editing and ~29.97% faster correcting versus touch-only, and ~30.81% and ~47.96% reductions versus iOS Voice Control.
AI-written summary · about
chapters (3)
related talks
12mTyping Haptically: Towards Enabling Non-auditory Smartphone Text Entry with Haptic Feedback for B...
20mHybrid-Brailler: Combining Physical and Gestural Interaction for Mobile Braille Input and Editing
21mVoicesetting: Voice Authoring UIs for Improved Expressivity in Augmentative Communication
10mTextFlow: Screenless Access to Non-Visual Smart Messaging
52mDesigning digital assistants for those who need it most
14mSmart Touch: Improving Touch Accuracy for People with Motor Impairments with Template Matching
55mDigital Modernization Through Accessibility
29mCan't touch this - Different ways users interact with their mobile devices