Update README.md

README.md

@@ -59,8 +59,28 @@ Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated sys
 
 
 ## How to Use the Model
-TODO
-
+The model is available for use in the NeMo toolkit [2], and can be used as a pre-trained checkpoint for inference.
+
+### Automatically load the model
+
+```python
+import nemo.collections.asr as nemo_asr
+asr_model = nemo_asr.models.EncDecFrameClassificationModel.from_pretrained(model_name="frame_vad_multilingual_marblenet_v2.0.nemo")
+```
+
+### Perform VAD Inference
+
+```bash
+python <NEMO_ROOT>/examples/asr/speech_classification/frame_vad_infer.py \
+  --config-path="../conf/vad" \
+  --config-name="frame_vad_infer_postprocess.yaml" \
+  vad.model_path=<Path to .nemo file from which model should be instantiated> \ 
+  input_manifest=<Path of manifest file of evaluation data, where audio files should have unique names> \
+  prepare_manifest.auto_split=True \
+  prepare_manifest.split_duration=7200 \
+  vad.parameters.shift_length_in_sec=0.02 \
+  out_manifest_filepath=<Path of output manifest file>
+```
 
 ## Software Integration:
 **Runtime Engine(s):**