We propose a lecture-on-demand system, which searches lecture videos for segments relevant to user information needs. We utilize the benefits of textbooks and audio/video data corresponding to a single lecture. Our system extracts the audio track from a target lecture video, generates a transcription by large vocabulary continuous speech recognition, and produces a textual index. Users can selectively view specific video segments by submitting textual queries associated with the textbook for the target lecture. Experimental results showed that by adapting speech recognition to the lecture topic, the recognition accuracy increased and the retrieval accuracy was comparable with that obtained by human transcriptions. Our system is implemented as a client-server system over the Web to facilitate e-education.