wav2vec2-base-checkpoint-10 Open-source Speech Recognition Model - Fine-tuned Based on Datasets, Low Error Rate

Wav2vec2 Base Checkpoint 10

Developed by jiobiala24

A speech recognition model fine-tuned on the common_voice dataset based on wav2vec2-base-checkpoint-9, achieving a word error rate of 0.3292 on the evaluation set

Speech Recognition

Transformers

Open Source License:Apache-2.0 #Speech Recognition #Low Word Error Rate #Multi-round Fine-tuning

Downloads 16

Release Time : 3/2/2022

Model Overview

This is a speech recognition model based on the wav2vec2 architecture, fine-tuned on the common_voice dataset, capable of converting speech to text.

Model Features

Low Word Error Rate

Achieved a word error rate of 0.3292 on the evaluation set, demonstrating good performance

Based on wav2vec2 Architecture

Utilizes the wav2vec2-base architecture, which has excellent speech feature extraction capabilities

Fine-tuning Optimization

Underwent 30 rounds of fine-tuning training on the common_voice dataset

Model Capabilities

Speech-to-Text

Automatic Speech Recognition

Use Cases

Speech Transcription

Convert speech content into written records

Word error rate 0.3292

Voice Assistants

Voice Command Recognition

Recognize user voice commands

Training Loss	Epoch	Step	Validation Loss	Wer
0.2892	1.62	1000	0.5745	0.3467
0.235	3.23	2000	0.6156	0.3423
0.1782	4.85	3000	0.6299	0.3484
0.1504	6.46	4000	0.6475	0.3446
0.133	8.08	5000	0.6753	0.3381
0.115	9.69	6000	0.7834	0.3529
0.101	11.31	7000	0.7924	0.3426
0.0926	12.92	8000	0.7887	0.3465
0.0863	14.54	9000	0.7674	0.3439
0.0788	16.16	10000	0.8648	0.3435
0.0728	17.77	11000	0.8460	0.3395
0.0693	19.39	12000	0.8941	0.3451
0.0637	21.0	13000	0.9079	0.3356
0.0584	22.62	14000	0.8851	0.3336
0.055	24.23	15000	0.9400	0.3338
0.0536	25.85	16000	0.9387	0.3335
0.0481	27.46	17000	0.9664	0.3337
0.0485	29.08	18000	0.9567	0.3292

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Wav2vec2 Base Checkpoint 10

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 wav2vec2-base-checkpoint-10

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License