Update README.md

README.md

@@ -147,6 +147,18 @@ model-index:
 This model transcribes speech in lower case English alphabet along with spaces and apostrophes.
 It is a "extra-large" versions of Conformer-Transducer (around 600M parameters) model.  
 
+## NVIDIA Riva: Deployment
+
+If you like this and other models from NVIDIA (i.e., CTC-based Conformers) check out [NVIDIA Riva](https://developer.nvidia.com/riva), an accelerated speech AI SDK deployable on-prem, in all clouds, multi-cloud, hybrid, on edge, and embedded. This model, as well as other RNNT-based models are currently not supported by Riva. You can find the list of models supported by Riva [here](https://docs.nvidia.com/deeplearning/riva/user-guide/docs/reference/models/index.html). 
+
+Additionally, Riva provides: 
+
+* World-class out-of-the-box accuracy for the most common languages with model checkpoints trained on proprietary data with hundreds of thousands of GPU-compute hours 
+* Best in class accuracy via customization with run-time word boosting (e.g., brand and product names), acoustic model training, language model training, and inverse text normalization customizations 
+* Streaming speech recognition, Kubernetes compatible scaling, and Enterprise-grade support 
+
+Check out [Riva live demo](https://developer.nvidia.com/riva#demos). 
+
 ## NVIDIA NeMo: Training
 
 To train, fine-tune or play with the model you will need to install [NVIDIA NeMo](https://github.com/NVIDIA/NeMo). We recommend you install it after you've installed latest Pytorch version.
@@ -154,10 +166,6 @@ To train, fine-tune or play with the model you will need to install [NVIDIA NeMo
 pip install nemo_toolkit['all']
 ``` 
 
-## NVIDIA Riva: Deployment
-
-[CTC-based Conformers](https://huggingface.co/nvidia/stt_en_conformer_ctc_large) are supported by Riva today. This model, as well as other RNNT-based models, will be supported by future versions of [NVIDIA Riva](https://developer.nvidia.com/riva). 
-
 ## How to Use this Model
 
 The model is available for use in the NeMo toolkit [3], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.