Knowledge transfer for Russian conversational telephone automatic speech recognition
نویسندگان
چکیده
منابع مشابه
Conversational telephone speech recognition
This paper describes the development of a speech recognition system for the processing of telephone conversations, starting with a state-of-the-art broadcast news transcription system. We identify major changes and improvements in acoustic and language modeling, as well as decoding, which are required to achieve state-of-theart performance on conversational speech. Some major changes on the aco...
متن کاملConfidence Annotation for Automatic Recognition of Conversational Telephone Speech
The construction and performance of a confidence annotator for a telephone channel speech recognizer is discussed. The confidence annotator labels each word output by the speech recognizer with a probability that indicates how certain it is that the output is correct. An application of this functionality is demonstrated in which the recognition system identifies automatically those conversation...
متن کاملImprovements in recognition of conversational telephone speech
This paper describes recent changes in Dragon’s speech recognition system which have markedly improved performance on conversational telephone speech. Key changes include: the conversion to modified PLP-based cepstra from mel-cepstra; the replacement of our usual IMELDA transformation by a new transform using “semi-tied covariance”; a new multi-pass adaptation protocol; probabilities on alterna...
متن کاملImproving English Conversational Telephone Speech Recognition
The goal of this work is to build a state-of-the-art English conversational telephone speech recognition system. We investigated several techniques to improve acoustic modeling, namely speaker-dependent bottleneck features, deep Bidirectional Long Short-Term Memory (BLSTM) recurrent neural networks, data augmentation and score fusion of DNN and BLSTM models. Training set consisted of the 300 ho...
متن کاملRecognition of conversational telephone speech using the JANUS speech engine
JANUS SPEECH ENGINE Torsten Zeppenfeld Michael Finke Klaus Ries Martin Westphal Alex Waibel Interactive Systems Laboratories Carnegie Mellon University, USA University of Karlsruhe, Germany ABSTRACT Recognition of conversational speech is one of the most challenging speech recognition tasks to-date. While recognition error rates of 10% or lower can now be reached on speech dictation tasks over ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Scientific and Technical Journal of Information Technologies, Mechanics and Optics
سال: 2018
ISSN: 2226-1494
DOI: 10.17586/2226-1494-2018-18-2-236-242