|
--- |
|
task_categories: |
|
- automatic-speech-recognition |
|
language: |
|
- en |
|
- fr |
|
- de |
|
- es |
|
tags: |
|
- music |
|
- lyrics |
|
- evaluation |
|
- benchmark |
|
- transcription |
|
pretty_name: 'JamALT: A Formatting-Aware Lyrics Transcription Benchmark' |
|
--- |
|
|
|
# JamALT: A Formatting-Aware Lyrics Transcription Benchmark |
|
|
|
JamALT is a revision of the [JamendoLyrics](https://github.com/f90/jamendolyrics) dataset, adapted for use as an automatic lyrics transcription (ALT) benchmark. |
|
|
|
The lyrics have been revised according to the newly compiled [annotation guidelines](GUIDELINES.md), which include rules about spelling, punctuation, and formatting. |
|
|
|
See the [project website](https://audioshake.github.io/jam-alt/) for details. |
|
|
|
## Loading the data |
|
|
|
```python |
|
from datasets import load_dataset |
|
dataset = load_dataset("audioshake/jam-alt")["test"] |
|
``` |
|
|
|
A subset is defined for each language (`en`, `fr`, `de`, `es`); |
|
for example, use `load_dataset("audioshake/jam-alt", "es")` to load only the Spanish songs. |
|
|
|
Other arguments can be specified to control audio loading: |
|
- `with_audio=False` to skip loading audio. |
|
- `sampling_rate` and `mono=True` to control the sampling rate and number of channels. |
|
- `decode_audio=False` to skip decoding the audio and just get the MP3 file paths. |
|
|
|
## Running evaluation |
|
|
|
Use the [`alt-eval`](https://github.com/audioshake/alt-eval) package for evaluation: |
|
```python |
|
from alt_eval import compute_metrics |
|
compute_metrics(dataset["text"], transcriptions, languages=dataset["language"]) |
|
``` |
