SpeechBrain

Open-source PyTorch toolkit for conversational AI, speech recognition, and speaker verification

Voice & AudioFree

SpeechBrain is an open-source, PyTorch-based toolkit for developing speech AI systems. It provides pre-built recipes and models for automatic speech recognition, speaker verification, speech enhancement, text-to-speech, language identification, and spoken language understanding. Researchers and AI engineers use SpeechBrain to build and fine-tune production speech AI systems without building from scratch. The toolkit's modular design allows mixing and matching components across tasks. Academic researchers use SpeechBrain for reproducing state-of-the-art results and conducting new research, while companies use it as the foundation for custom speech AI products requiring full control over the model training pipeline.

Key Features

• ✓Speech recognition
• ✓Speaker verification
• ✓Speech enhancement
• ✓TTS
• ✓Open-source PyTorch