The paper describes a hardware-software system for real-time closed captioning of Russian live TV broadcasts. The use of respeaking technology enabled us to create an ASR system with WER not exceeding 5.5%. Editing closed captions in real time further reduces WER down to 0.2%. In the paper we report some advancements in LMs for a highly inflected language and also in using morphological rescoring of the decoder word lattice. We propose a solution of the punctuation problem and effective methods of real-time editing of ASR results. This system was successfully used during paralympic games in Sochi for live web-broadcasting on russiasport.ru. We are reporting work in progress and are planning to achieve even better ASR accuracy in the course of the next year.
Bibliographic reference. Levin, K. / Ponomareva, I. / Bulusheva, A. / Chernykh, G. / Medennikov, I. / Merkin, N. / Prudnikov, A. / Tomashenko, Natalia (2014): "Automated closed captioning for Russian live broadcasting", In INTERSPEECH-2014, 1438-1442.