Update README.md

README.md

@@ -19,7 +19,7 @@ tags:
 # Model Card for Model ID
 ## Model Details
 ### Model Description
-The model consists of a music encoder ```MERT-v1-300M```, a natural language decoder ```vicuna-7b-delta-v0```, and a linear projection laer between the two.
+The model consists of a music encoder ```MERT-v1-300M```, a natural language decoder ```vicuna-7b-delta-v0```, and a linear projection layer between the two.
 
 This checkpoint of MusiLingo is developed on the MusicQA and can answer instructions with music raw audio, such as querying about the tempo, emotion, genre, tags or subjective feelings etc.
 You can use the MusicQA dataset for the following demo. For the implementation of MusicQA, please refer to our [Github repo](https://github.com/zihaod/MusiLingo/blob/main/musilingo/datasets/datasets/musicqa_dataset.py).