Update README.md

README.md

@@ -16,8 +16,8 @@ tags:
 
 ## Model Overview
 
-In order to prepare, adjust, or experiment with the model, it's necessary to install NVIDIA NeMo.
-We advise installing it once you've already installed the most recent version of Pytorch.
+In order to prepare, adjust, or experiment with the model, it's necessary to install NVIDIA NeMo Toolkit [1].
+We advise installing it once you've installed the most recent version of Pytorch.
 ```
 pip install nemo_toolkit['all']
 ```
@@ -47,13 +47,36 @@ python3 transcribe_speech.py model_path=stt_kz_quartznet15x5.nemo dataset_manife
 
 ## Input and Output
 
-This model can take input in the form of mono-channel audio .WAV files with a sample rate of 
-16,000 KHz. Then, this model gives you the spoken words in a text format for a given audio sample.
+This model can take input from mono-channel audio .WAV files with a sample rate of 16,000 KHz. 
+Then, this model gives you the spoken words in a text format for a given audio sample.
 
 ## Model Architecture
 
-QuartzNet [2] is a Jasper-like network that uses separable convolutions and larger filter sizes. It has comparable accuracy to Jasper while having much fewer parameters. This particular model has 15 blocks each repeated 5 times.
+QuartzNet 15x5 [2] is a Jasper-like network that uses separable convolutions and larger filter sizes. It has comparable accuracy to Jasper while having much fewer parameters. This particular model has 15 blocks each repeated 5 times.
 
+## Training
 
+The model was finetuned to Kazakh speech based on the pre-trained English Model for over several epochs.
 
+## Dataset
 
+Kazakh Speech Corpus 2 (KSC2) [3] is the first industrial-scale open-source Kazakh speech corpus.
+In total, KSC2 contains around 1.2k hours of high-quality transcribed data comprising over 600k utterances.
+
+## Performance
+
+Average WER: 15.53%
+
+## Limitation
+
+Because the GPU (NVIDIA GeForce RTX 2070) has limited power, we used a lightweight model architecture for fine-tuning. 
+In general, this makes it faster for inference but might show less overall performance. 
+In addition, if the speech includes technical terms or dialect words the model hasn't learned, it may not work as well.
+
+## References
+
+[1] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)
+
+[2] [QuartzNet 15x5](https://catalog.ngc.nvidia.com/orgs/nvidia/models/quartznet15x5)
+
+[3] [Kazakh Speech Corpus 2](https://issai.nu.edu.kz/kz-speech-corpus/?version=1.1)