my_awesome_minds_model Open-source Speech Recognition Model - Accurately recognize speech based on the minds14 dataset

My Awesome Minds Model

Developed by stevhliu

A speech recognition model fine-tuned on the minds14 dataset based on facebook/wav2vec2-base

Speech Recognition

Transformers

Open Source License:Apache-2.0 #Speech Classification #wav2vec2 Fine-tuning #Low Accuracy Model

Downloads 107

Release Time : 10/7/2022

Model Overview

This model is a fine-tuned model for speech recognition tasks, based on the wav2vec2 architecture and trained on the minds14 dataset.

Model Features

Based on wav2vec2 Architecture

Uses facebook's wav2vec2-base as the foundational architecture, with excellent speech feature extraction capabilities

Fine-tuned on minds14 Dataset

Specially optimized and trained for the minds14 dataset

Model Capabilities

Speech Recognition

Audio Classification

Use Cases

Speech Processing

Voice Command Recognition

Can be used to recognize simple voice commands

🚀 my_awesome_minds_model

This model is a fine - tuned version of [facebook/wav2vec2 - base](https://huggingface.co/facebook/wav2vec2 - base) on the minds14 dataset. It helps in achieving certain performance metrics on speech - related tasks, providing a valuable tool for relevant applications.

🚀 Quick Start

This model is a fine - tuned version of [facebook/wav2vec2 - base](https://huggingface.co/facebook/wav2vec2 - base) on the minds14 dataset. It achieves the following results on the evaluation set:

Loss: 2.6986
Accuracy: 0.0442

📚 Documentation

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e - 05
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.8	3	2.6691	0.0708
No log	1.8	6	2.6824	0.0442
No log	2.8	9	2.6804	0.0708
3.1934	3.8	12	2.6884	0.0796
3.1934	4.8	15	2.6903	0.0531
3.1934	5.8	18	2.6876	0.0442
3.1711	6.8	21	2.6911	0.0531
3.1711	7.8	24	2.6977	0.0442
3.1711	8.8	27	2.6991	0.0442
3.152	9.8	30	2.6986	0.0442