Speech Corpus Creation Tool
Support by Parquet, CSV, Jsonl, XLS
Create Reddit dataset
Provide feedback on AI responses to prompts
Build datasets using natural language
Perform OSINT analysis, fetch URL titles, fine-tune models
Generate a Parquet file for dataset validation
Label data for machine learning models
Browse and view Hugging Face datasets from a collection
Launch and explore labeled datasets
Clean and process datasets
Browse and extract data from Hugging Face datasets
Build datasets using natural language
Dhravani is a Speech Corpus Creation Tool designed to help users create high-quality speech datasets by recording voices and transcribing them. It simplifies the process of building speech corpora, making it accessible for both researchers and developers.
• User-Friendly Interface: Designed for ease of use, allowing users to record and transcribe audio seamlessly.
• High-Quality Audio Recording: Ensures clear and accurate voice recordings for better dataset quality.
• Automatic Transcription: Converts recorded audio into text, saving time and effort.
• Data Organization: Manages recorded audio and transcriptions in an organized structure for easy access.
• Export Capabilities: Allows users to export datasets in various formats for further analysis or model training.
What is a speech corpus?
A speech corpus is a collection of speech data used to train and test speech recognition systems or other AI models.
Can I use Dhravani for multiple languages?
Yes, Dhravani supports multiple languages, allowing users to create diverse speech datasets.
Is my recorded data secure?
Dhravani ensures that all recordings and transcriptions are stored securely, with options for encryption and privacy protection.