As more and more online search queries come from voice, automatic speechrecognition becomes a key component to deliver relevant search results . Errors introduced by automatic speech recognition (ASR) lead to irrelevant search results returned to the user, thus causing user dissatisfaction . In this paper, we introduce an approach, Mondegreen, to correct voice queries in text spacewithout depending on audio signals, which may not always be available due to constraints or privacy or bandwidth (for example, some ASR systems run on-device) considerations . We see the approach ascomplementing existing highly-optimized productionASR systems, which can not be frequently retrained and thus lag behind due to vocabulary drifts. We then demonstrate that Mondegreens can achieve significantimprovements in increased user interaction by correcting user voice queries . We then demonstrated that Mondegareen can

Author(s) : Sukhdeep S. Sodhi, Ellie Ka-In Chio, Ambarish Jash, Santiago Ontañón, Ajit Apte, Ankit Kumar, Ayooluwakunmi Jeje, Dima Kuzmin, Harry Fung, Heng-Tze Cheng, Jon Effrat, Tarush Bali, Nitin Jindal, Pei Cao, Sarvjeet Singh, Senqiang Zhou, Tameen Khan, Amol Wankhede, Moustafa Alzantot, Allen Wu, Tushar Chandra

Links : PDF - Abstract

Code :
Coursera

Keywords : search - queries - voice - user - systems -

Leave a Reply

Your email address will not be published. Required fields are marked *