This study proposes a method to detect problematic speech segments automatically for second language (L2) listeners, considering both lexical and acoustic aspects. It introduces a tool, Partial and Synchronized Caption (PSC), which provides assistance for language learners and fosters L2 listening skills. PSC presents purposively selected words along with a video in the form of synchronized text-to-speech captions. It uses corpus-based information and conducts speech-data analysis to detect difficulties in speech, thus achieving effective word selection. To this end, PSC uses an Automatic Speech Recognition (ASR) system as a model for L2 listeners to elucidate speech difficulties for these learners. In this method, misrecognized words by the ASR system are analyzed to find useful patterns that could signal problematic speech segments for L2 listeners. The identified patterns were evaluated by experiments to ensure that they cause difficulties for L2 listeners in the same way that they impede ASR performance. Experimental findings confirm that adding these instances to PSC significantly improves learners' recognition of the respective segments. [For the complete volume, see ED578177.]