Evaluation of interface and in-car speech - many undesirable utterances and sever noisy speech on car navigation application -

Abstract

In this paper, we report two evaluation results of ASR (Automatic Speech Recognition) for a car navigation system, first interface evaluation results using a commercial product and second ASR module evaluation results for noisy in-car speech. In the first evaluation, we report many undesirable OOV (Out Of Vocabulary) utterances which make the interface worse. To overcome this problem, we propose a sophisticated interface which can handle OOV problems. Next, we carried out ASR module evaluation for the noisy in-car real speech gathered using driving car. We used free Continuous Speech Recognition (CSR) software Julius/Julian as a speech decoder.

Usability evaluation of a Volkswagen Group in-vehicle speech system

“…The problem areas are system-wide and not specific to any particular task, and often spanned across multiple tasks. Problem areas we identified are comparable to those reported in other studies of speech system usability [3,4,5]. Those problem areas that were observed in 5 or more participants are listed in the table below (see Table 3).…”

“…These challenges range from issues with the sophistication of the speech technology to the user interaction schemes used to guide task performance. With respect to the speech technology itself, issues such as car noise interference and a limited vocabulary of speech commands that the system can recognize [5] constrain the ways that humans can interact with the system. On the usability side, many speech interfaces do not have a clear and transparent menu structure [5], which leads to confusion about why a certain command is misrecognized in certain contexts but not others.…”

“…With respect to the speech technology itself, issues such as car noise interference and a limited vocabulary of speech commands that the system can recognize [5] constrain the ways that humans can interact with the system. On the usability side, many speech interfaces do not have a clear and transparent menu structure [5], which leads to confusion about why a certain command is misrecognized in certain contexts but not others. Also, the pace of speech system dialogues is far from the natural pace of human conversation [8].…”

A Secondary Assessment of the Impact of Voice Interface Turn Delays on Driver Attention and Arousal in Field Conditions

“…Long delay times may also be due to hardware constraints or voice recognition issues resulting from environmental noise (Hataoka et al, 2008). Regardless, selfreport data show that people find long delays uncomfortable and unnatural (Skantze & Hjalmarsson, 2012;Meena, Skantze, & Gustafson 2014).…”