ORAL READING ASSESSMENT

Reading is an essential skill to become successful in any discipline. Oral Reading Assessment app provides a way for the students' to improve their reading skills. At present, the application supports indian english and hindi languages. Students are required to read the text with clear pronouncation. Then students can see their transcribed audio, and other key metrics under the results section.

Setting up the project

1. Clone the repo

Clone this repository for all the source code as mentioned below:

$ git clone https://github.com/PlaypowerLabs/Oral-Reading-Assessment.git
$ cd Oral-Reading Assessment

2. Installating Packages & Downloading Models

• The application is made using Streamlit in Python. Make sure that least version of python is 3.4.0 .

• Use the below command to download all the python libraries and some dependencies required for the application to run:

$ pip install --no-cache-dir -r requirements.txt
$ python -m install spacy download en_core_web_sm

• Download models (Total Size - 2.3GB) using the below command:

$ chmod +x data_download.sh
$ ./data_download.sh

This would create models folder inside the Oral-Reading-Assessment repo. To manually download the data, click the below link:

https://oral-reading-assessment-app-models.s3.us-east-2.amazonaws.com/models.zip

Running the Oral Reading Assessment Webapp

• Execute the python file app.py using streamlit:

$ streamlit run app.py

• The webapp is launched in your browser and opened automatically.

• Choose the language that you want to practice speaking and click Show Passage button.

• Wait for the speech-to-text engine to load for the selected language and then click Record button and start reading the passage.

• Under the results section you can view your transcribed audio and some key metrics for assessing the reading ability like WER, WPM and Confidence.