Real-time audio recognition protocol

A computer system configured to process an audio sample, the computer system comprising a non-transitory computer-readable medium encoding instructions thereon that, in response to execution by the computer system, configure the computer system to:

• receive, at the computer system from a client device via a first network connection, a first window of an audio sample, the audio sample being derived from an audio recording made by the client device, and the first window comprising less than an entirety of the audio sample;
• determine a first recognition result corresponding to the first window, the first recognition result having a first content type;
• and
• provide, by the computer system to the client device via a second network connection, the first recognition result prior to receiving the entirety of the audio sample from the client device.

A computer-implemented method for processing an audio sample comprising:

• establishing, by a client device, a first hypertext transfer protocol connection with a server;
• establishing, by the client device, a second hypertext transfer protocol connection with the server;
• recording, by the client device, an audio sample;
• sending, at the client device to the server via the first hypertext transfer connection protocol, a first window of the audio sample, the first window comprising less than an entirety of the audio sample;
• and
• receiving, at the client device from the server via the second hypertext transfer protocol connection, a first recognition result corresponding to the first window, the first recognition result being received prior to the client device sending the entirety of the audio sample to the server.