S

Stt Bm Quartznet15x5 V0

Developed by RobotsMali
This is a Bambara automatic speech recognition model fine-tuned based on the NVIDIA NeMo framework, suitable for Bambara speech-to-text tasks.
Downloads 88
Release Time : 2/7/2025

Model Overview

This model is a fine-tuned version of NVIDIA stt_fr_quartznet15x5, optimized specifically for Bambara automatic speech recognition and trained using the CTC loss function.

Model Features

Bambara Optimization
Specially fine-tuned and optimized for Bambara speech recognition
Lightweight Architecture
Uses the QuartzNet 15x5 architecture with only 19M parameters, suitable for resource-limited environments
Continuous Improvement
Part of an ongoing research project, with further optimizations planned for future versions

Model Capabilities

Bambara speech recognition
16kHz mono audio processing

Use Cases

Speech-to-Text
Bambara Speech Transcription
Convert Bambara speech into text
Achieved a WER of 46.5% on the test set
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase