Spaces:
Running
on
Zero
Running
on
Zero
Update README.md
Browse filesdocs: Update Readme
README.md
CHANGED
@@ -11,9 +11,11 @@ pinned: false
|
|
11 |
|
12 |
# πΌΒ SonicVerse
|
13 |
|
14 |
-
An interactive demo for SonicVerse
|
15 |
that includes a general description of the music as well as music features such as key, instruments, genre, mood / theme, vocals gender.
|
16 |
|
|
|
|
|
17 |
---
|
18 |
|
19 |
## π Demo
|
@@ -25,7 +27,8 @@ Check out the live Space here:
|
|
25 |
|
26 |
## π Samples
|
27 |
|
28 |
-
Short captions
|
|
|
29 |
|
30 |
---
|
31 |
|
@@ -35,6 +38,10 @@ Short captions
|
|
35 |
|
36 |
β
Upload a long music clip (upto 1 minute for successful demo) to get a long detailed caption for the whole music clip.
|
37 |
|
|
|
|
|
|
|
|
|
38 |
---
|
39 |
|
40 |
## π οΈ How to Run Locally
|
@@ -57,19 +64,6 @@ python app.py
|
|
57 |
|
58 |
---
|
59 |
|
60 |
-
<!-- ## π File Structure
|
61 |
-
|
62 |
-
```
|
63 |
-
.
|
64 |
-
βββ app.py # Web app file
|
65 |
-
βββ requirements.txt # Python dependencies
|
66 |
-
βββ environment.yml # Conda environment
|
67 |
-
βββ README.md # This file
|
68 |
-
βββ src/sonicverse # Source
|
69 |
-
```
|
70 |
-
|
71 |
-
--- -->
|
72 |
-
|
73 |
## π‘ Usage
|
74 |
|
75 |
To use the app:
|
@@ -79,6 +73,31 @@ To use the app:
|
|
79 |
|
80 |
---
|
81 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
82 |
## π§Ή Built With
|
83 |
|
84 |
- [Hugging Face Spaces](https://huggingface.co/spaces)
|
@@ -87,14 +106,3 @@ To use the app:
|
|
87 |
- [MERT 95M](https://huggingface.co/m-a-p/MERT-v1-95M)
|
88 |
---
|
89 |
|
90 |
-
<!-- ## β¨ Acknowledgements
|
91 |
-
|
92 |
-
- [Model authors or papers you built on]
|
93 |
-
- [Contributors or collaborators]
|
94 |
-
|
95 |
-
---
|
96 |
-
|
97 |
-
## π License
|
98 |
-
|
99 |
-
This project is licensed under the MIT License / Apache 2.0 / Other.
|
100 |
-
-->
|
|
|
11 |
|
12 |
# πΌΒ SonicVerse
|
13 |
|
14 |
+
An interactive demo for **SonicVerse**, a music captioning model, allowing users to input audio and generate a natural language caption
|
15 |
that includes a general description of the music as well as music features such as key, instruments, genre, mood / theme, vocals gender.
|
16 |
|
17 |
+
The demo supports both short (10s) and long (up to 1 minute) audio inputs.
|
18 |
+
|
19 |
---
|
20 |
|
21 |
## π Demo
|
|
|
27 |
|
28 |
## π Samples
|
29 |
|
30 |
+
Short captions and long chained LLM-generated captions:
|
31 |
+
β‘οΈ [Samples page](https://amaai-lab.github.io/SonicVerse/)
|
32 |
|
33 |
---
|
34 |
|
|
|
38 |
|
39 |
β
Upload a long music clip (upto 1 minute for successful demo) to get a long detailed caption for the whole music clip.
|
40 |
|
41 |
+
β
Captions include musical attributes (key, instruments, tempo, etc.)
|
42 |
+
|
43 |
+
β οΈ You can upload audio of any length, but due to compute limitations on Hugging Face Spaces, we recommend uploading clips under **30 seconds** unless you have a **Hugging Face Pro account** or run the app locally.
|
44 |
+
|
45 |
---
|
46 |
|
47 |
## π οΈ How to Run Locally
|
|
|
64 |
|
65 |
---
|
66 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
67 |
## π‘ Usage
|
68 |
|
69 |
To use the app:
|
|
|
73 |
|
74 |
---
|
75 |
|
76 |
+
## π Citation
|
77 |
+
|
78 |
+
If you use SonicVerse in your work, please cite our paper:
|
79 |
+
|
80 |
+
**SonicVerse: Multi-Task Learning for Music Feature-Informed Captioning**
|
81 |
+
Anuradha Chopra, Abhinaba Roy, Dorien Herremans
|
82 |
+
Accepted to AIMC 2025
|
83 |
+
|
84 |
+
```bibtex
|
85 |
+
@article{chopra2025sonicverse,
|
86 |
+
title={SonicVerse: Multi-Task Learning for Music Feature-Informed Captioning},
|
87 |
+
author={Chopra, Anuradha and Roy, Abhinaba and Herremans, Dorien},
|
88 |
+
journal={Proceedings of the 6th Conference on AI Music Creativity (AIMC 2025)},
|
89 |
+
year={2025},
|
90 |
+
address={Brussels, Belgium},
|
91 |
+
month={September},
|
92 |
+
url={https://arxiv.org/abs/2506.15154},
|
93 |
+
}
|
94 |
+
```
|
95 |
+
|
96 |
+
Read the paper here: [arXiv:2506.15154](https://arxiv.org/abs/2506.15154)
|
97 |
+
DOI: [10.48550/arXiv.2506.15154](https://doi.org/10.48550/arXiv.2506.15154)
|
98 |
+
|
99 |
+
---
|
100 |
+
|
101 |
## π§Ή Built With
|
102 |
|
103 |
- [Hugging Face Spaces](https://huggingface.co/spaces)
|
|
|
106 |
- [MERT 95M](https://huggingface.co/m-a-p/MERT-v1-95M)
|
107 |
---
|
108 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|