Mid-span device and system

1. A peripheral video processing device, comprising:

a plurality of first interfaces to communicably connect the peripheral video device to a camera as a plug in, said first interfaces configured to receive one or more videos from the camera;

a second interface configured to receive audio from the camera;

a unit configured to split videos into multiple streams;

a pass-through processing unit configured to send the original stream as an output without any latency;

a module to receive one or more user inputs corresponding to one or more customization requests to customize the one or more videos and audio received;

one or more processing modules to make one or more customizations to one or more split videos and audio based on the one or more customization requests, wherein the one or more processing modules are configured to determine at least one of an active speaker in one or more received videos and a number of faces in the one or more received videos; and

an output interface to output customized video and audio.

2. The video processing device of claim 1 , wherein the one or more customizations comprises zooming in on the one or more videos.

3. The video processing device of claim 1 , wherein the one or more customizations comprises creating a composite video of the received one or more videos.

4. The video processing device of claim 1 , wherein the one or more videos are asynchronously received.

5. The video processing device of claim 1 , further comprising a plurality of buffers to store each of the received one or more videos.

6. The video processing device of claim 1 , wherein the one or more customizations comprises converting the received one or more videos into a H.264 video format.

7. The video processing device of claim 1 , wherein the one or more customizations comprises outputting a passthrough of the audio received.

8. The video processing device of claim 1 , wherein the output video corresponds to the video including the determined active speaker.

9. The video processing device of claim 1 , wherein the output interface outputs a plurality of videos simultaneously.

10. The video processing device of claim 1 , configured to recognize faces in the one or more received videos.

11. The video processing device of claim 10 , configured to recognize speech corresponding to the determined active speaker and generate one or more transcripts of the recognized speech.

12. The video processing device of claim 11 , wherein the generated one or more transcripts are associated with an ID of the active speaker.

13. A video processing method comprising:

receiving one or more videos through a plurality of first interfaces at a peripheral device plugged in to a camera;

receiving, at the peripheral device, an audio via a second interface;

processing, by the peripheral device, the one or more videos and the audio based on the received one or more user inputs comprising splitting the videos into multiple streams and sending, by a pass-through unit the original video stream as an output without any latency;

receiving, at the peripheral device, one or more user inputs corresponding to one or more customization requests corresponding to the one or more videos received and the audio received;

customizing, by the peripheral device, the one or more videos and the audio based on the received one or more user inputs comprising determining at least one of an active speaker in one or more received videos and a number of faces in the one or more received videos; and

outputting, by the peripheral device, the customized video and audio via an output interface.