Flush Barrier

Accumulates all audio and events during runtime. Releases them in the order they were received after Flush is called.
Parameters:
None.

Inputs

Audio:
Accepts Everything

Events:
Accepts Everything

Outputs

Audio:
Writes all the incoming audio to output as is after Flush is received.

Events:
Writes all the incoming events to output as is after Flush is received.

Remarks :

• This node releases every accumulated audio and event after Flush is called. So until all the input is provided and the flush method is called this node does not output anything. Hence it is only suitable for batch usage.

Project Structure
A sample project can be built as such:

The aim of this project is to only perform transcriptions after Flush is called.
Taking this a step further, we can create a project like:

In this project, all audio segments wait until Language Identifier outputs every Language Change event. Making sure the SR Http node does recognitions in the Last Identified Language.
We can create a more complex project such as:

In this project, Language Identifier only outputs one Language Change event. The Audio Segment Picker node makes sure that the best segment is used for Language Identification.
This project is used to guarantee that the SR Http recognitions are performed in the Most Probable Language.

Supported flow types: Batch