A method, computer system, and a computer program product for playing a video recording that includes captions, wherein each caption includes a textual transcription of an audio portion of the video, a description of non-speech elements of the video, or both, is provided. The present invention may include rendering a video, identifying a first portion of the video that includes a first caption, and, in response to identifying the first portion of the video, estimating a second time period for a particular user to read and understand text in the first caption. The present invention may further include determining whether the second time period for the particular user to read and understand text in the first caption is greater than the first time period for rendering the first portion of the video. The present invention may lastly include pausing the video for a third time period.