The open-source speech recognition model, english - filipino - wav2vec2 - l - xls - r - test - 08, accurately recognizes Filipino speech!

English Filipino Wav2vec2 L Xls R Test 08

Developed by Khalsuu

This is a speech recognition model fine-tuned on Filipino speech datasets based on the wav2vec2-large-xlsr-53-english model

Speech Recognition

Transformers

Open Source License:Apache-2.0 #Filipino Speech Recognition #Multilingual Speech Transcription #Low Word Error Rate

Downloads 25

Release Time : 5/7/2022

Model Overview

This model is used for English-to-Filipino speech recognition tasks, achieving a word error rate (WER) of 0.4255 on the evaluation set

Model Features

Multilingual Support

Supports speech recognition in both English and Filipino

Efficient Fine-tuning

Fine-tuned based on the pre-trained wav2vec2-large-xlsr-53 model

Low Word Error Rate

Achieved a word error rate (WER) of 0.4255 on the evaluation set

Model Capabilities

English Speech Recognition

Filipino Speech Recognition

Speech-to-Text Conversion

Use Cases

Speech Transcription

English-Filipino Speech Transcription

Convert English or Filipino speech content into text

Word Error Rate 0.4255

Multilingual Applications

Bilingual Voice Assistant

Voice interaction system supporting both English and Filipino

Training Loss	Epoch	Step	Validation Loss	Wer
3.3434	2.09	400	2.2857	0.9625
1.6304	4.19	800	1.1547	0.7268
0.9231	6.28	1200	1.0252	0.6186
0.6098	8.38	1600	0.9371	0.5494
0.4922	10.47	2000	0.7092	0.5478
0.3652	12.57	2400	0.7358	0.5149
0.2735	14.66	2800	0.6270	0.4646
0.2038	16.75	3200	0.5717	0.4506
0.1552	18.85	3600	0.5968	0.4255

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 08

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 english-filipino-wav2vec2-l-xls-r-test-08

🚀 Quick Start

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

📄 License