Configuring smart-rec-start-time in DeepStream Graph

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.3
• TensorRT Version 8.6.1.6-1+cuda12.0
• NVIDIA GPU Driver Version (valid for GPU only) 535.104.12

• Issue Type( questions, new requirements, bugs) questions

I am currently utilizing the Graph Composer tool to construct an application that incorporates multiple input sources with smart recording functionality enabled. I am interested in configuring the smart-rec-start-time parameter, but I am unsure of the correct procedure.

According to the documentation, I am instructed to add this parameter to the “source” section. So I added it under the multiple_source_input section of the graph YAML file, shown below.

components:
- name: multiple_source_input
  parameters:
    audio-%u-out: audio_output
    smart-rec-cache: 20
    smart-rec-start-time: 5  ####
    smart-rec-container: 0
    smart-rec-default-duration: 0
    ...

However, when executing the graph, I encountered an error:

WARN gxf/std/yaml_file_loader.cpp@1058: Using unregistered parameter ‘smart-rec-start-time’ in component ‘multiple_source_input’

My inquiries are as follows:

  1. Does DeepStream 6.3 support this particular configuration?
  2. What is the correct method for configuring this parameter for multiple source input?

Please use the composer UI to configure the correct extension and the configurations.

Thank you for the reply.

I checked the multi-source input properties, shown below, but smart-rec-start-time is not included… am I missing something?

The NvDsMultiSrcInputWithRecord is implemented based on Gst-nvurisrcbin — DeepStream documentation 6.4 documentation

“smart-rec-start-time” is only for smart recording initialization, the “smart-rec-cache” is the actual parameter which impacts the cached video/audio.