diff --git a/JOSS/paper.md b/JOSS/paper.md
index 826f592..e414428 100644
--- a/JOSS/paper.md
+++ b/JOSS/paper.md
@@ -27,6 +27,8 @@ bibliography: paper.bib
 ---
 
 # Summary:
+PyBCI is a comprehensive, open-source Python framework developed to facilitate brain-computer interface (BCI) research. It encompasses data acquisition, data labelling, feature extraction, and machine learning. PyBCI provides a streamlined, user-friendly platform for creating real-time BCI applications. The software uses the Lab Streaming Layer (LSL) [@lsl] protocol for the unified collection of time-series measurement that handles both the networking, time-synchronization and (near-) real-time access (supported LSL devices found here: https://labstreaminglayer.readthedocs.io/info/supported_devices.html). At least one LSL data stream is required and a single marker stream is used for labelling training data. PyTorch [@NEURIPS2019_9015], SciKit-learn [@scikit-learn] and TensorFlow [@tensorflow2015-whitepaper] are leveraged for applying machine learning classifiers. NumPy [@oliphant2006guide], SciPy [@2020SciPy-NMeth], and Antropy [@vallat_antropy_2023] are utilised for generic time and/or frequency feature extraction examples.
+
 
 PyBCI is a comprehensive, open-source Python framework developed to facilitate brain-computer interface (BCI) research. It encompasses data acquisition, data labelling, feature extraction, and machine learning. PyBCI provides a streamlined, user-friendly platform for conducting real-time BCI applications. The software uses the Lab Streaming Layer (LSL) [@lsl] protocol for data acquisition on various LSL enabled data streams. The LSL is a system for the unified collection of measurement time series in research experiments that handles both the networking (supported LSL devices found here: https://labstreaminglayer.readthedocs.io/info/supported_devices.html). At least one LSL data stream is required and a single marker stream is used for labelling training data. PyTorch [@NEURIPS2019_9015], SciKit-learn [@scikit-learn] and TensorFlow [@tensorflow2015-whitepaper] are leveraged for applying machine learning classifiers. NumPy [@oliphant2006guide], SciPy [@2020SciPy-NMeth], and Antropy [@vallat_antropy_2023] are utilised for generic time and/or frequency feature extraction examples.