4. Model Artifacts and Setup

The AM62DX EdgeAI image includes three pre-compiled, pre-packaged models ready to run out of the box — no compilation required. These models, artifacts, and setup steps are common to both the Linux inference application and the GStreamer pipelines.

4.1. Pre-Packaged Models

Model	Type	ModelId	Artifacts path	Description
GCRN	Speech enhancement	2	`/usr/share/tvm_inference/artifacts/gcrn/`	Removes background noise from speech; produces enhanced WAV output
YAMNet	Audio classification	4	`/usr/share/tvm_inference/artifacts/yamnet/`	521-class AudioSet audio classification
VGGish	Audio classification	3	`/usr/share/tvm_inference/artifacts/vggish/`	10-class UrbanSound8K audio classification

4.2. Model Artifacts

Each model directory under /usr/share/tvm_inference/artifacts/<model>/ contains:

File	Description
`deploy_graph.json`	Graph description; defines the neural network topology and input/output shapes
`deploy_lib.so`	Compiled model library; shared object for TVM runtime
`deploy_param.params`	Model weights; binary parameter data

4.3. Label Files

Classification models ship with label files under /usr/share/tvm_inference/labels/:

File	Description
`yamnet_label_list.txt`	521 lines; one AudioSet class name per line for YAMNet
`vggish_label_list.txt`	10 lines; one UrbanSound8K class name per line for VGGish

4.4. Input Audio

All pipelines expect 16 kHz, mono, signed 16-bit PCM WAV input. Sample files in this format are pre-installed under /usr/share/tvm_inference/input/:

File	Used with	Description
`input_audio.wav`	GCRN	Sample noisy speech recording
`speech_whistling2.wav`	YAMNet	Sample audio for event classification
`urbansound_16k.wav`	YAMNet, VGGish	Sample urban sound recording

If your source audio is in a different format, convert it before use:

ffmpeg -i input.wav -ar 16000 -ac 1 -sample_fmt s16 output_16k.wav

Or convert inline inside a GStreamer pipeline by inserting audioconvert and audioresample before the audio/x-raw capsfilter.

4.5. Pre-Packaged Libraries

The following libraries are pre-installed on the AM62DX EdgeAI image. No installation is required to run the out-of-box pipelines on the AM62D EVM.

Library	Description
`libti_rpmsg_dma`	RPMsg and DMA client library for inter-processor communication
`libtvm_runtime.so`	Apache TVM ARM64 runtime for model inference
`libgstti.so`	GStreamer plugin library containing `tidspkernel` and `titvm` elements
`libasound`	ALSA audio library
`libsndfile`	Audio file I/O library
`libjson-c`	JSON parsing library

4.6. Model Daemon

The tvm_model_daemon keeps a TVM model loaded in memory and serves inference requests over a Unix domain socket at /var/run/tvm-inference.sock. Both the Linux inference application and the GStreamer titvm element connect to it automatically, eliminating repeated artifact loading for each run.

Start the daemon:

root@am62dxx-evm:~# systemctl enable --now tvm-model-daemon.service
root@am62dxx-evm:~# journalctl -u tvm-model-daemon -f
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: rpmsg: Opened endpoint device /dev/rpmsg4
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: [daemon] Loading TVM artifacts from: /usr/share/tvm_inference/artifacts/gcrn
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: [daemon] C7x firmware OK and already running
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: c7x: Connected to compute service
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: c7x: Staging buffer:  0xffff6c89e000 (phys 0xa3000000)
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: c7x: Result buffer: 0xffff7549e000 (phys 0xabc00000)
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: [08:05:48] {TVM_HOME}/src/runtime/contrib/tidl/tidl_runtime_rpmsg.cc:216: TVM AM62D: ping status=0 version=0x20000 uptime=17438ms
Oct 23 08:05:48 am62dxx-evm tvm_model_daemon[997]: [08:05:48] {TVM_HOME}/src/runtime/contrib/tidl/tidl_runtime_rpmsg.cc:239: TVM AM62D: loading deploy_mod (51527424 bytes) with 1 input name(s)
Oct 23 08:05:49 am62dxx-evm tvm_model_daemon[997]: c7x: Loaded module handle=1 (text=0 data=0 input_offset=51527443)
Oct 23 08:05:49 am62dxx-evm tvm_model_daemon[997]: [08:05:49] {TVM_HOME}/src/runtime/contrib/tidl/tidl_runtime_rpmsg.cc:263: TVM AM62D: module loaded, handle=1

To preload artifacts at boot:

root@am62dxx-evm:~# systemctl enable --now tvm-model-preload.service

If the daemon is not running, both the Linux application and GStreamer titvm element will wait up to 30 seconds before falling back to loading the model in-process.

4.7. DSP Firmware Check

Both the Linux application and GStreamer pipelines require the C7x™ DSP firmware to be running. Verify before use:

root@am62dxx-evm:~# cat /sys/class/remoteproc/remoteproc0/state
running