This Audio lectures and speeches contain a
wealth of valuable information, but reviewing and
extracting the key points can be tedious and time-
consuming. This paper presents an automated system
that uses speech recognition and text summarization
techniques to identify and summarize the most salient
content from spoken presentations. Audio is first
transcribed to text via a speech recognition engine. The
resulting text is then processed by an extractive
summarization algorithm based on term frequency-
inverse document frequency (TF-IDF) to extract the most
important points. These summarized points can
optionally be used to generate relevant supplementary
URLs that provide additional context or resources related
to the topics covered. This system was developed to
enable quick review of lectures and speeches by
automatically delivering condensed, relevant summaries.
Keywords : Speech Recognition, Extractive Summarization, TF-IDF, URLs.