🚀 VoxLingua107 ECAPA-TDNN Spoken Language Identification Model
This is a spoken language recognition model that can classify speech utterances according to the spoken language. It covers 107 different languages and is trained on the VoxLingua107 dataset using SpeechBrain, offering high - performance language identification capabilities.
✨ Features
- Multilingual Support: Capable of classifying 107 different languages, including Abkhazian, Afrikaans, Amharic, etc.
- Automatic Audio Normalization: Automatically normalizes audio (resampling + mono channel selection) when calling
classify_file if needed.
- Dual - Purpose Usage: Can be used 'as is' for spoken language recognition or as an utterance - level feature (embedding) extractor.
📦 Installation
pip install git+https://github.com/speechbrain/speechbrain.git@develop
💻 Usage Examples
Basic Usage
import torchaudio
from speechbrain.inference.classifiers import EncoderClassifier
language_id = EncoderClassifier.from_hparams(source="speechbrain/lang-id-voxlingua107-ecapa", savedir="tmp")
signal = language_id.load_audio("speechbrain/lang-id-voxlingua107-ecapa/udhr_th.wav")
prediction = language_id.classify_batch(signal)
print(prediction)
print(prediction[1].exp())
print(prediction[3])
emb = language_id.encode_batch(signal)
print(emb.shape)
Advanced Usage
To perform inference on the GPU, add run_opts={"device":"cuda"} when calling the from_hparams method.
language_id = EncoderClassifier.from_hparams(source="speechbrain/lang-id-voxlingua107-ecapa", savedir="tmp", run_opts={"device":"cuda"})
📚 Documentation
Model description
This is a spoken language recognition model trained on the VoxLingua107 dataset using SpeechBrain.
The model uses the ECAPA - TDNN architecture that has previously been used for speaker recognition. However, it uses
more fully connected hidden layers after the embedding layer, and cross - entropy loss was used for training.
We observed that this improved the performance of extracted utterance embeddings for downstream tasks.
The system is trained with recordings sampled at 16kHz (single channel).
The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling classify_file if needed.
Intended uses & limitations
The model has two uses:
- Use 'as is' for spoken language recognition
- Use as an utterance - level feature (embedding) extractor, for creating a dedicated language ID model on your own data
The model is trained on automatically collected YouTube data. For more
information about the dataset, see here.
Limitations and bias
Since the model is trained on VoxLingua107, it has many limitations and biases, some of which are:
- Probably it's accuracy on smaller languages is quite limited
- Probably it works worse on female speech than male speech (because YouTube data includes much more male speech)
- Based on subjective experiments, it doesn't work well on speech with a foreign accent
- Probably it doesn't work well on children's speech and on persons with speech disorders
Training data
The model is trained on VoxLingua107.
VoxLingua107 is a speech dataset for training spoken language identification models.
The dataset consists of short speech segments automatically extracted from YouTube videos and labeled according the language of the video title and description, with some post - processing steps to filter out false positives.
VoxLingua107 contains data for 107 languages. The total amount of speech in the training set is 6628 hours.
The average amount of data per language is 62 hours. However, the real amount per language varies a lot. There is also a seperate development set containing 1609 speech segments from 33 languages, validated by at least two volunteers to really contain the given language.
Training procedure
See the SpeechBrain recipe.
Evaluation results
Error rate: 6.7% on the VoxLingua107 development dataset
Referencing
Referencing SpeechBrain
@misc{speechbrain,
title={{SpeechBrain}: A General-Purpose Speech Toolkit},
author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
year={2021},
eprint={2106.04624},
archivePrefix={arXiv},
primaryClass={eess.AS},
note={arXiv:2106.04624}
}
Referencing VoxLingua107
@inproceedings{valk2021slt,
title={{VoxLingua107}: a Dataset for Spoken Language Recognition},
author={J{\"o}rgen Valk and Tanel Alum{\"a}e},
booktitle={Proc. IEEE SLT Workshop},
year={2021},
}
About SpeechBrain
SpeechBrain is an open - source and all - in - one speech toolkit. It is designed to be simple, extremely flexible, and user - friendly. Competitive or state - of - the - art performance is obtained in various domains.
Website: https://speechbrain.github.io/
GitHub: https://github.com/speechbrain/speechbrain
🔧 Technical Details
- Model Architecture: Uses the ECAPA - TDNN architecture, with more fully connected hidden layers after the embedding layer.
- Loss Function: Cross - entropy loss is used for training.
- Audio Sampling: The system is trained with recordings sampled at 16kHz (single channel).
📄 License
This model is licensed under the "apache - 2.0" license.
⚠️ Important Note
In the dataset and in the defaults of this model (see label_encoder.txt, the used ISO language code for Hebrew is obsolete (should be he instead of iw). The ISO language code for Javanese is incorrect (should be jv instead of jw). See issue #2396.
💡 Usage Tip
The system is trained with recordings sampled at 16kHz (single channel). The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling classify_file if needed. Make sure your input tensor is compliant with the expected sampling rate if you use encode_batch and classify_batch.
| Property |
Details |
| Model Type |
Spoken Language Identification Model |
| Training Data |
VoxLingua107 |
| Metrics |
Accuracy |
| License |
apache - 2.0 |
| Datasets |
VoxLingua107 |
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
Transformers