first

README.md

@@ -1,5 +1,19 @@
+---
+license: mit
+tags:
+  - small audio-language model
+  - ALM
+  - audio
+  - music
+  - sound events
+  - audio reasoning
+  - audio captioning
+  - audio question answering
+  - zero-shot
+  - audio-text
+---
 # Mellow
-[[`Paper`]()] [[`Checkpoint`]()]
+[[`Paper`]()] [[`GitHub`](https://github.com/soham97/Mellow)] [[`Checkpoint`](https://huggingface.co/soham97/Mellow)]
 
 Mellow is a small Audio-Language Model that takes in two audios and a text prompt as input and produces free-form text as output. It is a 167M parameter model and trained on ~155 hours of audio (AudioCaps and Clotho), and achieves SoTA performance on different tasks with 50x fewer parameters. 
 
@@ -79,7 +93,7 @@ print(f"\noutput: {response}")
 The composition of the ReasonAQA dataset is shown in Table. The training set is restricted to AudioCaps and Clotho audio files and the testing is performed on 6 tasks - Audio Entailment, Audio Difference, ClothoAQA, Clotho MCQ, Clotho Detail, AudioCaps MCQ and AudioCaps Detail.
 
 ![alt text](resource/data.png)
-- The ReasonAQA JSONs can be downloaded from Zenodo: [checkpoint \[drive\]](https://drive.google.com/file/d/1WPKgafYw2ZCifElEtHn_k3DkcVGjesqB/view?usp=sharing)
+- The ReasonAQA JSONs can be downloaded from Zenodo: [checkpoint](https://drive.google.com/file/d/1WPKgafYw2ZCifElEtHn_k3DkcVGjesqB/view?usp=sharing)
 - The audio files can be downloaded from their respective hosting website: [Clotho](https://zenodo.org/records/4783391) and [AudioCaps](https://github.com/cdjkim/audiocaps)
 
 ## Limitation