metadata
license: mit
language:
- as
base_model:
- openai/whisper-medium
Assamese Dialect Classification Model
This repository contains a trained model for classifying Assamese dialects based on speech inputs. The model was developed to assist in identifying and understanding regional variations of the Assamese language.
Model Purpose
The purpose of this model is to classify different dialects of Assamese speech. It is useful for linguistic research, speech analysis, and creating dialect-aware applications in natural language processing (NLP) and automatic speech recognition (ASR).
Dialects Recognized
The model is trained to recognize the following four dialects of the Assamese language:
- Darangia
- Kamrupia
- Nalbaria
- Upper Assam
Training Dataset
The model was trained on a dataset of 300 speech samples, curated to include diverse speakers, phrases, and dialect features. The dataset includes:
- Diverse Data: Various accents, speaker genders, and age groups.
- Metadata: Information about speaker age, gender, district, and speech duration.
- Common Phrases: Speech samples based on frequently used phrases in Assamese.