Open, simple, flexible, well-documented, and with competitive performance.
SpeechBrain supports state-of-the-art technologies for speech recognition, enhancement, separation, text-to-speech, speaker recognition, speech-to-speech translation, spoken language understanding, and beyond.
SpeechBrain encompasses a wide range of audio technologies, including vocoding, audio augmentation, feature extraction, sound event detection, beamforming, and other multi-microphone signal processing capabilities.
SpeechBrain offers user-friendly tools for training Language Models, supporting technologies ranging from basic n-gram LMs to modern Large Language Models. Our platform seamlessly integrates them into speech processing pipelines and facilitates the creation of customizable chatbots.
SpeechBrain leverages the most advanced deep learning technologies, including methods for self-supervised learning, continual learning, diffusion models, Bayesian deep learning, and interpretable neural networks.
SpeechBrain is engineered to accelerate the research and development of Conversational AI technologies. It comes with pre-built recipes for popular datasets. Extensive documentation and tutorials are available to support newcomers.
SpeechBrain offers pre-trained models with user-friendly interfaces, making tasks like transcription, speaker verification, speech enhancement, and source separation easier than ever.
# From PyPI
pip install speechbrain
# Local installation
git clone https://github.com/speechbrain/speechbrain.git
cd speechbrain
pip install -r requirements.txt
pip install --editable .
cd recipes/{dataset}/{task}/train
# Train the model using the default recipe
python train.py hparams/train.yaml
# Train the model with a hyperparameter tweak
python train.py hparams/train.yaml --learning_rate=0.1
class ASR_Brain(sb.Brain):
def compute_forward(self, batch, stage):
# Compute features (mfcc, fbanks, etc.) on the fly
features = self.hparams.compute_features(batch.wavs)
# Improve robustness with pre-built augmentations
features = self.hparams.augment(features)
# Apply your custom model
return self.modules.myCustomModel(features)
