updated

README.md

@@ -58,13 +58,10 @@ asr_model = nemo_asr.models.ASRModel.from_pretrained("iqbalc/stt_de_conformer_tr
 ```
 
 ### Transcribing using Python
-First, let's get a sample
 ```
-wget https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wav
+simply do:
 ```
-Then simply do:
-```
-asr_model.transcribe(['2086-149220-0033.wav'])
+asr_model.transcribe(['filename.wav'])
 ```
 
 ### Transcribing many audio files
@@ -83,32 +80,32 @@ This model provides transcribed speech as a string for a given audio sample.
 
 ## Model Architecture
 
-<ADD SOME INFORMATION ABOUT THE ARCHITECTURE>
+Conformer-Transducer model is an autoregressive variant of Conformer model for Automatic Speech Recognition which uses Transducer loss/decoding
 
 ## Training
 
-<ADD INFORMATION ABOUT HOW THE MODEL WAS TRAINED - HOW MANY EPOCHS, AMOUNT OF COMPUTE ETC>
+The NeMo toolkit was used for training the models. These models are fine-tuned with this example script and this base config.
+
+The tokenizers for these models were built using the text transcripts of the train set with this script.
 
 ### Datasets
 
-<LIST THE NAME AND SPLITS OF DATASETS USED TO TRAIN THIS MODEL (ALONG WITH LANGUAGE AND ANY ADDITIONAL INFORMATION)>
+All the models in this collection are trained on a composite dataset comprising of over two thousand hours of cleaned German speech:
+
+1. MCV7.0 567 hours 
+2. MLS 1524 hours 
+3. VoxPopuli 214 hours
 
 ## Performance
 
-<LIST THE SCORES OF THE MODEL - 
-      OR
-USE THE Hugging Face Evaluate LiBRARY TO UPLOAD METRICS>
+Performances of the ASR models are reported in terms of Word Error Rate (WER%) with greedy decoding.
 
-## Limitations
+MCV7.0 test	= 4.93
 
-<DECLARE ANY POTENTIAL LIMITATIONS OF THE MODEL>
+## Limitations
 
-Eg: 
-Since this model was trained on publically available speech datasets, the performance of this model might degrade for speech which includes technical terms, or vernacular that the model has not been trained on. The model might also perform worse for accented speech.
+The model might perform worse for accented speech
 
 
 ## References
-
-<ADD ANY REFERENCES HERE AS NEEDED>
-
-[1] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)
+[NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)