Developed a bot-side silence gap insertion feature for the pipecat-ai/pipecat repository, enhancing audio processing by ensuring that bot utterances are clearly segmented during recording. Leveraging Python and asynchronous programming, the implementation introduced a dedicated gap-filling method and precise timing tracking to mirror user-side gap handling, thereby reducing audio concatenation artifacts and improving speech clarity. The work included expanding unit test coverage with eight new tests to validate the updated bot-side behavior, strengthening end-to-end conversation capture. This focused engineering effort addressed a nuanced audio integrity issue, resulting in more accurate and maintainable handling of idle periods in conversation recordings.