Open-source model wav2vec2-xls-r-300m-ur-cv9-with-lm - Precise automatic Urdu speech recognition

Wav2vec2 Xls R 300m Ur Cv9 With Lm

Developed by anuragshas

This model is an automatic speech recognition (ASR) model fine-tuned on Urdu speech datasets based on facebook/wav2vec2-xls-r-300m

Speech Recognition

Transformers

OtherOpen Source License:Apache-2.0 #Urdu speech recognition #Low word error rate #Multilingual support

Downloads 18

Release Time : 5/4/2022

Model Overview

This is a speech recognition model optimized for Urdu, trained on the Common Voice 9.0 Urdu dataset, capable of converting Urdu speech to text

Model Features

Urdu optimization

Specially fine-tuned for Urdu speech recognition tasks

High performance

Achieves a word error rate (WER) of 23.75% and a character error rate (CER) of 8.31% on the Common Voice test set

Based on XLS-R architecture

Utilizes Facebook's powerful XLS-R (Cross-lingual Speech Representation) architecture

Model Capabilities

Urdu speech recognition

Speech-to-text

Automatic transcription

Use Cases

Speech transcription

Urdu speech transcription

Convert Urdu speech content to text

Accuracy rate of 76.25% (WER 23.75%)

Voice assistants

Urdu voice command recognition

For Urdu voice assistants and command control systems

🚀 XLS-R-300M - Urdu

This is a fine - tuned model for automatic speech recognition on the Urdu dataset, achieving good performance in evaluation.

🚀 Quick Start

This model is a fine - tuned version of [facebook/wav2vec2 - xls - r - 300m](https://huggingface.co/facebook/wav2vec2 - xls - r - 300m) on the MOZILLA - FOUNDATION/COMMON_VOICE_9_0 - UR dataset. It achieves the following results on the evaluation set:

Loss: 0.4147
Wer: 0.3172
Cer: 0.1050

✨ Features

Task Type: Automatic Speech Recognition
Datasets: mozilla - foundation/common_voice_9_0
Metrics: wer

📚 Documentation

Model Index

Property	Details
Name	XLS - R - 300M - Urdu
Task Type	Automatic Speech Recognition
Dataset Type	mozilla - foundation/common_voice_9_0
Dataset Name	Common Voice 9
Dataset Args	ur
Metrics - WER	23.750 (Test WER)
Metrics - CER	8.310 (Test CER)

Training Procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 7.5e - 05
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 2
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
training_steps: 5108
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
3.2894	7.83	400	3.1501	1.0	1.0
1.8586	15.68	800	0.8871	0.6721	0.2402
1.3431	23.52	1200	0.5813	0.5502	0.1939
1.2052	31.37	1600	0.4956	0.4788	0.1665
1.1097	39.21	2000	0.4447	0.4143	0.1397
1.0528	47.06	2400	0.4439	0.3961	0.1333
0.9939	54.89	2800	0.4348	0.4014	0.1379
0.9441	62.74	3200	0.4236	0.3653	0.1223
0.913	70.58	3600	0.4309	0.3475	0.1157
0.8678	78.43	4000	0.4270	0.3337	0.1110
0.8414	86.27	4400	0.4158	0.3220	0.1070
0.817	94.12	4800	0.4185	0.3231	0.1072

Framework versions

Transformers 4.19.0.dev0
Pytorch 1.11.0+cu102
Datasets 2.1.1.dev0
Tokenizers 0.12.1

📄 License

This model is licensed under the Apache - 2.0 license.

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