Facebook, Inc.

Mathieu R Henaire

Amit Puntambekar

Sonal Gandhi

Michael Hamilton Coward

Minchuan Chen

Sabyasachi Roy

An online system receives live stream content to be provided to one or more client devices as the live stream is ongoing. The online system generates a variety of transcoded live stream content to ensure that client devices can readily playback the appropriately encoded content. Once the live stream ends, the online system determines whether to use a transcoded live stream content or to use the original content received by the online system in order to generate a video on demand. To do so, the online system considers the quality of the original content and issues associated with the original content such as missing video frames or missing audio excerpts. The online system can choose to normalize the original content by eliminating the issues to improve the quality of the normalized original content and subsequently generate the video on demand from the normalized original content.

Live to video on demand normalization

An online system receives a media stream from a host of a live broadcasting stream and a second media stream from a guest of the host, and generates a single composite stream that includes the multiple live streams. The online system decodes and aligns the video and audio frames of each live stream to ensure that the composite stream displays the multiple live streams in a synchronized fashion. Additionally, the composite stream can display the multiple live streams in a particular format (e.g. side-by-side, or picture-in-picture), and therefore, the online system adjusts the video and audio frames of each live stream to fit that format. For each composite stream, the online system employs a decision engine that identifies the optimal set of encoded composite streams based on a variety of characteristics. The online system encodes the composite stream and distributes the appropriately encoded composite stream to client devices for display to viewers.

Presentation of composite streams to users

An online system receives content (e.g. video content and/or audio content) from a content provider device to be appropriately transcoded and distributed to client devices for consumption. During the transcoding process, the online system decodes the content into an intermediate format, and. considers a variety of characteristics to decide the optimal set of content encodings for encoding the decoded content. Characteristics include source characteristics associated with the content provider device, social graph characteristics of the user of the online system that is using the content provider device, viewer characteristics associated with the client devices that request to access the content, and content characteristics. The online system encodes the content for each client device with appropriate encoders selected based on the optimal set of content encodings. For each client device that requests to access the content, the online system provides encoded content that the client device can appropriately playback.

Decision engine for dynamically selecting media streams

In one embodiment, a computer server machine generates a time map associated with the video stream. The video stream comprises one or more time slices. The computer server machine identifies a non-relevant time frame. The computer machine receives a comment and a localized comment time stamp, and associates the comment with one or more time slices. One or more client devices communicates a request for one or more time slices to the computer server machine, and the computer server machine determines whether the localized comment time stamp is within the requested one or more time slices. When the localized comment time stamp is within the requested one or more time slices, computer server machine transforms the localized time stamp to a synchronized comment time stamp. Computer server machine sends information regarding the synchronized comment time stamp to the one or more client devices.

Comment synchronization in a video stream

In one embodiment, a method includes receiving a series of a series of audio-and-video segment pairs in a live video feed. The series of audio-and-video segment pairs is traversed, comparing the end-time of a current audio-and-video segment with the start-time of the next audio-and-video segment pair, in the sequence, and identifying any time gap between consecutive audio-and-video segment pairs. When a time gap is identified, the end-time of a current segment pair is subtracted from the start time of the next segment-pair to define an offset time, and subsequent start times of segment pairs in the sequence are adjusted based on this offset time. If after adjusting a start time of the next segment-pair by the offset time, a gap in the audio segment remains, then the gap is filled with an encoded silence sequence.

Growing community of inventors

Redmond, WA, United States of America

Mathieu R Henaire