Can't configure & get meta from Gst-nvinferaudio module in python app

[juan.isern95]

Hi everyone

• Hardware Platform (Jetson / GPU): RTX2080TI
• DeepStream Version: 6.2
• TensorRT Version: 8.5.2-1+cuda11.8
• NVIDIA GPU Driver Version (valid for GPU only): 530.30.02
• Issue Type( questions, new requirements, bugs): Question

I have been trying to integrate my audio classification model based on resnet18 into Deestream. However, while I think the whole pipeline makes sense (I think so), I am not able to pick up the meta associated with that PGIE. My processing flow is as follows.

My nvaudioinfer definition file is as follows:

definition.txt (3.1 KB)

Checking the Gstreamer logs with LEVEL=3 I get the following:

0:00:03.196596303 21 0x7fd4ed7ff0f0 WARN nvinferaudio gstnvinferaudio.cpp:282:gst_nvinferaudio_start: error: Failed to create audio transform

I have tried to set this property in the PGIE definition.txt file and in code. But I can’t get it to work.

It is worth mentioning that as input I use an RTSP video stream with audio.

Best regards,

Juan Isern

[fanzh]

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

please refer to depstream audo sample \opt\nvidia\deepstream\deepstream-6.2\sources\apps\sample_apps\deepstream-audio\deepstream_audio_main.c
from the log, “audio-transform” is not set, please refer to \opt\nvidia\deepstream\deepstream-6.2\sources\apps\sample_apps\deepstream-audio\configs\ds_audio_sonyc_test_config.txt