Create TechnicalNotes
Browse filesAdd initial Technical Notes
- docs/TechnicalNotes +23 -0
docs/TechnicalNotes
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# OpenTunes AI Technical Notes
|
2 |
+
|
3 |
+
## Architecture
|
4 |
+
|
5 |
+
OpenTunes AI uses a combination of natural language processing (NLP) and machine learning (ML) to generate music. The platform is built using the following technologies:
|
6 |
+
|
7 |
+
* Python 3.9.7
|
8 |
+
* Transformers 4.17.0
|
9 |
+
* Gradio 3.0.2
|
10 |
+
|
11 |
+
## Model Architecture
|
12 |
+
|
13 |
+
The music generation model used in OpenTunes AI is a variant of the T5 model, which is a text-to-text transformer model. The model is fine-tuned on a dataset of music-related text and audio files.
|
14 |
+
|
15 |
+
## Training Data
|
16 |
+
|
17 |
+
The training data for the music generation model consists of a dataset of music-related text and audio files. The dataset is curated from a variety of sources, including music websites and APIs.
|
18 |
+
|
19 |
+
## Future Work
|
20 |
+
|
21 |
+
* Improve the quality of the generated music by fine-tuning the model on a larger dataset.
|
22 |
+
* Add support for multiple music genres and styles.
|
23 |
+
* Develop a more user-friendly interface for the Gradio app.
|