English-Filipino Wav2Vec2-L-XLS-R Test-04 Open-Source Model - Freely Deployable for English-Filipino Speech Recognition

English Filipino Wav2vec2 L Xls R Test 04

Developed by Khalsuu

This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-english on the filipino_voice dataset, designed for English-Filipino speech recognition tasks.

Speech Recognition

Transformers

Open Source License:Apache-2.0 #Filipino speech recognition #Multilingual transfer learning #Low-resource optimization

Downloads 21

Release Time : 5/6/2022

Model Overview

This is a speech recognition model based on the wav2vec2 architecture, specifically optimized for mixed English and Filipino speech.

Model Features

Multilingual support

Supports speech recognition for both English and Filipino

Based on wav2vec2 architecture

Utilizes the advanced wav2vec2-large-xlsr-53 architecture with excellent speech recognition performance

Fine-tuned optimization

Specially fine-tuned on Filipino speech datasets, suitable for Filipino language environments

Model Capabilities

English speech recognition

Filipino speech recognition

Mixed-language speech recognition

Use Cases

Speech-to-text

Filipino speech transcription

Convert Filipino speech content into text

Word Error Rate (WER) of 0.5078

English-Filipino mixed speech processing

Process mixed speech content containing both English and Filipino

Training Loss	Epoch	Step	Validation Loss	Wer
2.2131	2.09	400	0.7100	0.6832
0.6539	4.19	800	0.8307	0.6602
0.5081	6.28	1200	0.7120	0.6297
0.42	8.38	1600	0.7309	0.6299
0.3482	10.47	2000	0.7665	0.6148
0.293	12.57	2400	0.7091	0.5840
0.265	14.66	2800	0.8170	0.6102
0.2294	16.75	3200	0.9715	0.6216
0.1872	18.85	3600	0.8516	0.5837
0.1644	20.94	4000	0.8408	0.5767
0.1495	23.04	4400	0.9188	0.5717
0.1276	25.13	4800	1.0149	0.5451
0.116	27.23	5200	1.0220	0.5683
0.1017	29.32	5600	0.9319	0.5253
0.0899	31.41	6000	0.9949	0.5435
0.0861	33.51	6400	1.1029	0.5467
0.0766	35.6	6800	1.0219	0.5193
0.065	37.7	7200	1.0836	0.5214
0.0588	39.79	7600	1.0713	0.5078

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

English Filipino Wav2vec2 L Xls R Test 04

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 english-filipino-wav2vec2-l-xls-r-test-04

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

🔧 Technical Details

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License