AMajor/musiclang-v2
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Florian GARDIN 2024-02-29 16:51:27 +00:00 committed by system
parent 052c94531e
commit 108f155a70
No known key found for this signature in database
GPG Key ID: 6A528E38E0733467
1 changed files with 123 additions and 161 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

284
README.md
View File

@ -3,199 +3,161 @@ library_name: transformers
tags: [] tags: []
--- ---
# Model Card for Model ID MusicLang : Controllable Symbolic Music Generation
========================================================
<!-- Provide a quick summary of what the model is/does. --> ![MusicLang logo](https://github.com/MusicLang/musiclang/blob/main/documentation/images/MusicLang.png?raw=true "MusicLang")
🎶 <b>&nbsp; You want to generate music that you can export to your favourite DAW in MIDI ?</b>
## Model Details
### Model Description 🎛️ <b>&nbsp; You want to control the chord progression of the generated music ? </b>
<!-- Provide a longer summary of what this model is. -->
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. 🚀 <b>&nbsp; You need to run it fast on your laptop without a gpu ?</b>
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional] Here is MusicLang Predict, your controllable music copilot.
<!-- Provide the basic links for the model. --> I just want to try !
--------------------
- **Repository:** [More Information Needed] [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1MA2mek826c05BjbWk2nRkVv2rW7kIU_S?usp=sharing)
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses Go to our Colab, we have a lot of cool examples. From generating creative musical ideas to continuing a song with a specified chord progression.
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> I am more serious about it
--------------------------
### Direct Use Install the musiclang-predict package :
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> ```bash
pip install musiclang_predict
```
[More Information Needed] Then open your favourite notebook and start generating music in a few lines :
### Downstream Use [optional] ```python
from musiclang_predict import MusicLangPredictor
nb_tokens = 1024
temperature = 0.9 # Don't go over 1.0, at your own risks !
top_p = 1.0 # <=1.0, Usually 1 best to get not too much repetitive music
seed = 16 # change here to change result, or set to 0 to unset seed
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> ml = MusicLangPredictor('musiclang/musiclang-v2') # Only available model for now
[More Information Needed] score = ml.predict(
nb_tokens=nb_tokens, # 1024 tokens ~ 25s of music (depending of the number of instruments generated)
temperature=temperature,
topp=top_p,
rng_seed=seed # change here to change result, or set to 0 to unset seed
)
score.to_midi('test.mid') # Open that file in your favourite DAW, score editor or even in VLC
```
### Out-of-Scope Use You were talking about controlling the chord progression ?
----------------------------------------------------------
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> You had a specific harmony in mind am I right ?
That's why we allow a fine control over the chord progression of the generated music.
Just specify it as a string like below, choose a time signature and let the magic happen.
[More Information Needed] ```python
from musiclang_predict import MusicLangPredictor
## Bias, Risks, and Limitations # Control the chord progression
# Chord qualities available : M, m, 7, m7b5, sus2, sus4, m7, M7, dim, dim0.
# You can also specify the bass if it belongs to the chord (eg : Bm/D)
chord_progression = "Am CM Dm E7 Am" # 1 chord = 1 bar
time_signature = (4, 4) # 4/4 time signature, don't be too crazy here
nb_tokens = 1024
temperature = 0.8
top_p = 1.0
seed = 42
<!-- This section is meant to convey both technical and sociotechnical limitations. --> ml = MusicLangPredictor('musiclang/musiclang-v2')
[More Information Needed] score = ml.predict_chords(
chord_progression,
time_signature=time_signature,
temperature=temperature,
topp=top_p,
rng_seed=seed # set to 0 to unset seed
)
score.to_midi('test.mid', tempo=120, time_signature=(4, 4))
```
### Recommendations Disclaimer : The chord progression is not guaranteed to be exactly the same as the one you specified. It's a generative model after all.
Usually it will happen when you use an exotic chord progression and if you set a high temperature.
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. That's cool but I have my music to plug in ...
------------------------------------------------
## How to Get Started with the Model Don't worry, we got you covered. You can use your music as a template to generate new music.
Let's continue some Bach music with a chord progression he could have used :
```python
from musiclang_predict import MusicLangPredictor
from musiclang_predict import corpus
Use the code below to get started with the model. song_name = 'bach_847' # corpus.list_corpus() to get the list of available songs
chord_progression = "Cm C7/E Fm F#dim G7 Cm"
nb_tokens = 1024
temperature = 0.8
top_p = 1.0
seed = 3666
[More Information Needed] ml = MusicLangPredictor('musiclang/musiclang-v2')
## Training Details score = ml.predict_chords(
chord_progression,
score=corpus.get_midi_path_from_corpus(song_name),
time_signature=(4, 4),
nb_tokens=1024,
prompt_chord_range=(0,4),
temperature=temperature,
topp=top_p,
rng_seed=seed # set to 0 to unset seed
)
### Training Data score.to_midi('test.mid', tempo=110, time_signature=(4, 4))
```
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
What's coming next ?
[More Information Needed] ---------------------
### Training Procedure We are working on a lot of cool features, some are already encoded in the model :
- A control over the instruments used in each bar and their properties (note density, pitch range, average velocity)
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> - Some performances improvements over the inference C script
- A faster distilled model for real-time generation that can be embedded in plugins or mobile applications
#### Preprocessing [optional] - An integration into a DAW as a plugin
- Some specialized smaller models depending on our user's needs
[More Information Needed]
How does that work ?
---------------------
#### Training Hyperparameters
If you want to learn more about how we are moving toward symbolic music generation, go to our [technical blog](https://musiclang.github.io/).
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> The tokenization, the model are described in great details.
#### Speeds, Sizes, Times [optional] We are using a LLAMA2 architecture (many thanks to Andrej Karpathy awesome [llama2.c](https://github.com/karpathy/llama2.c)), trained on a large dataset of midi files (The CC0 licensed [LAKH](https://colinraffel.com/projects/lmd/)).
We heavily rely on preprocessing the midi files to get an enriched tokenization that describe chords & scale for each bar.
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> The is also helpful for normalizing melodies relative to the current chord/scale.
[More Information Needed]
Contributing & Contact us
## Evaluation -------------------------
<!-- This section describes the evaluation protocols and provides the results. --> We are looking for contributors to help us improve the model, the tokenization, the performances and the documentation.
If you are interested in this project, open an issue, a pull request, or even [contact us directly](https://www.musiclang.io/contact).
### Testing Data, Factors & Metrics
License
#### Testing Data -------
<!-- This should link to a Dataset Card if possible. --> Specific licenses applies to our models. If you would like to use the model in your product, please
[contact us](https://www.musiclang.io/contact). We are looking forward to hearing from you !
[More Information Needed] MusicLang Predict is licensed under the GPL-3.0 License.
#### Factors The MusicLang base language package on which the model rely ([musiclang package](https://github.com/musiclang/musiclang)) is licensed under the BSD 3-Clause License.
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]