The Domain Specific - NeMo Automatic Speech Recognition (ASR) Application facilitates training, evaluation, and performance comparison of ASR models. This NeMo application enables you to train or fine-tune pre-trained ASR models with your own data. Through this application, we empower you to create your own ASR models built for your domain-specific data. Developers have complete control over their data, unlike when using a “black box” ASR tool available in the cloud, giving you the ability to create better-performing ASR models for your use case.

Features

• Dataset Preparation and Exploration: Download pre-trained models, prepare datasets, and create projects for tracking.
• Acoustic Model Training and Fine-Tuning: Train and fine-tune acoustic models with your own data.
• Transcription (Speech to Text): Convert speech to text with the models created.
• Performance Tracking: Easily understand the performance progression of models.
• Model Export: Prepare models for deployment.

Benefits

• Enhanced Customization: Create ASR models tailored to your specific domain data.
• Improved Performance: Gain better performance compared to generic ASR tools.
• Complete Control: Maintain control over your data and model training process.
• Ease of Use: The application is packaged and easy to use, providing an end-to-end ASR system.
• Scalability: Suitable for various use cases, including cross-language fine-tuning and domain adaptation.