W

Wav2vec2 Xls R 1b Polish

Developed by jonatasgrosman
This is a Polish automatic speech recognition (ASR) model fine-tuned based on the XLS-R 1-billion parameter model, trained on datasets such as Common Voice 8.0, supporting 16kHz sampling rate audio input.
Downloads 212
Release Time : 3/2/2022

Model Overview

This model is an optimized automatic speech recognition system for Polish, fine-tuned from Facebook's XLS-R 1-billion parameter model, excelling in Polish speech recognition tasks.

Model Features

High-performance Polish recognition
Achieves 11.01% WER and 2.55% CER on the Common Voice 8.0 test set
Supports language model enhancement
With a language model, WER can be reduced to 7.32% and CER to 1.95%
Large-scale pre-training foundation
Fine-tuned from the XLS-R 1-billion parameter model, featuring powerful speech feature extraction capabilities
Multi-dataset training
Trained using Common Voice 8.0, Multilingual LibriSpeech, and Voxpopuli datasets

Model Capabilities

Polish speech recognition
16kHz audio processing
Batch speech transcription

Use Cases

Speech transcription
Speech-to-text services
Convert Polish speech content into text
Achieves 92.68% accuracy on standard test sets (with language model)
Voice assistants
Polish voice command recognition
Used for voice-controlled devices and applications
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase