wav2vec2-xls-r-300m-cv8-es Open-source Speech Recognition Model - Accurately Identify General Speech Scenarios

Wav2vec2 Xls R 300m Cv8 Es

Developed by edugp

A speech recognition model fine-tuned on the Common Voice dataset based on facebook/wav2vec2-xls-r-300m

Downloads 23

Release Time : 3/2/2022

Model Overview

This is a pretrained model for Spanish speech recognition, fine-tuned based on the wav2vec2-xls-r-300m architecture

Efficient Speech Recognition

Fine-tuned on the Common Voice dataset with high speech recognition accuracy

Large-scale Pretraining

Pretrained on the 300M-parameter wav2vec2-xls-r architecture

Multilingual Support

Specialized for Spanish speech recognition tasks

Spanish Speech Recognition

Audio-to-Text Conversion

Speech Processing

Speech Transcription

Spanish Speech Transcription

Convert Spanish speech content into text

Word Error Rate 0.1931

Voice Assistants

Spanish Voice Command Recognition

Used for voice command recognition in Spanish voice assistant systems

Property	Details
learning_rate	0.0003
train_batch_size	16
eval_batch_size	8
seed	42
gradient_accumulation_steps	2
total_train_batch_size	32
optimizer	Adam with betas=(0.9,0.999) and epsilon=1e - 08
lr_scheduler_type	linear
lr_scheduler_warmup_steps	500
num_epochs	30
mixed_precision_training	Native AMP

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base