Projects

Current work

• Remeeting: searchable conversations
• Mod9 ASR Engine: self-hosted speech recognition
• Developing tools to analyze body-worn camera footage (at Stanford)

Past projects

• Privacy Protecting Conversation Recorder (DHS SBIR)
• Discriminative acoustic features for deployable speech recognition (PhD)
• Knowledge-aided interface for Big Data streams (OSD/ARL SBIR)
• Rapidly deployable, low-resource ASR/KWS for IARPA's Babel Program
• Speech-transcript alignment service for online video captioning
• Won a prize at the Evernote Developer Competition
• Co-founded Modista (it didn't happen due to software patents)
• Speech recognition applications for the Parallel Computing Laboratory
• Mandarin and Arabic ASR for the DARPA GALE program
• VTLN for the English version of the SmartWeb recognizer
• Searching broadcast news video for the TREC Video Retrieval Evaluation
• Aligning imperfect transcripts of the Scottish Parliament (at CSTR)
• Speech recognition of non-native speakers (won an award)
• Graduate Student Instructor for CS188: Artificial Intelligence