Filipino - wav2vec2 - l - xls - r - 300m - test open-source speech recognition model accurately recognizes Filipino.

Filipino Wav2vec2 L Xls R 300m Test

Developed by Khalsuu

This model is a speech recognition model fine-tuned on the filipino_voice dataset based on facebook/wav2vec2-xls-r-300m, supporting Filipino language.

Speech Recognition

Transformers

Open Source License:Apache-2.0 #Filipino speech recognition #wav2vec2 fine-tuning #low-resource speech processing

Downloads 5,738

Release Time : 4/22/2022

Model Overview

This is a speech recognition model optimized for Filipino language, fine-tuned based on the wav2vec2-xls-r-300m architecture, and can be used to convert Filipino speech into text.

Model Features

Filipino Optimization

Specifically fine-tuned for Filipino speech characteristics to improve recognition accuracy

Based on XLS-R Architecture

Uses facebook's wav2vec2-xls-r-300m pre-trained model as the foundation, with powerful speech feature extraction capabilities

Relatively Lightweight

300M parameter size maintains performance while being relatively lightweight

Model Capabilities

Filipino speech recognition

Speech-to-text

Use Cases

Speech Transcription

Filipino Meeting Minutes

Automatically convert Filipino meeting recordings into text transcripts

Word Error Rate 0.4831

Voice Assistant

Provide speech recognition capabilities for Filipino voice assistants

Training Loss	Epoch	Step	Validation Loss	Wer
2.7314	2.09	400	0.7541	0.7262
0.6065	4.19	800	0.6738	0.6314
0.4063	6.28	1200	0.6310	0.5992
0.2986	8.38	1600	0.6301	0.5340
0.2263	10.47	2000	0.6598	0.5391
0.1714	12.57	2400	0.7778	0.5593
0.1303	14.66	2800	0.7231	0.4907
0.1056	16.75	3200	0.8031	0.4885
0.0851	18.85	3600	0.7753	0.4831

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

Filipino Wav2vec2 L Xls R 300m Test

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 Filipino-wav2vec2-l-xls-r-300m-test

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

🔧 Technical Details

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License