Browse by topic
Type of publication
Computer-Assisted Handwritten Text Transcription Using Speech Recognition. V Jornadas en Tecnología del Habla (VJTH'2008), 2008. pp. 229-232.Handwritten Text Recognition (HTR) has gained a lot of attention during the last few years. On the one hand, there exists a large number of applications that can benefit from it. On the other hand, HTR is similar to Automatic Speech Recognition (ASR). Thus, modelling techniques can be easily adapted and most of the already available ASR tools can be used to train and evaluate HTR systems quite straightforwardly. However, HTR performance is still far from perfect and post-editing is needed. This post-editing can be made efficiently by means of Computer Assisted Transcription of Handwritten Text Images (CATTI) tools, which iteratively interact with the user to achieve the desired transcription. Typically, this interaction has been made via mouse and keyboard. In this paper, we present the development of a new CATTI system which uses speech as an additional mean of interaction. The system is build upon a generic recogniser both for speech and handwriting recognition.