Update README.md

COVID-away CNNs/README.md

@@ -13,7 +13,7 @@ The COVID-away CNNs, once deployed on smartwatches, can instantly warn the users
  - The first model, "COVID-away_CNN_Acc.h5" requires only Accelerometer data (Acc = 39 features) to detect hand-to-face movement. 
  - The second model, "COVID-away_CNN_Acc+Gyro.h5" was trained using additional Gyroscope data (Acc+Gyro = 78 features), hence requiring real-time data from two sensors for hand-to-face movement detection. 
 
-In order to produce a smartwatch friendly version of the above models, we optimize the CNNs by quantizing both its weights & activations to INT-8. The latency & size optimized versions of above models are:
+In order to produce a smartwatch friendly version of the above models, we optimize the CNNs by quantizing both its weights & activations to INT-8. The latency and size optimized versions of above models are:
 
 - COVID-away_CNN_Acc_tflite_quantized.tflite
 - COVID-away_CNN_Acc+Gyro_tflite_quantized.tflite