Original datasets used for curation of BIGOS have specific terms of usage that must be understood and agreed to before use. Below are the links to the license terms and datasets the specific license type applies to:

Creative Commons 0 which applies to Common Voice
Creative Commons By Attribution Share Alike 4.0, which applies to Clarin Cyfry, Azon acoustic speech resources corpus.
Creative Commons By Attribution 3.0, which applies to CLARIN Mobile database, CLARIN Studio database, PELCRA Spelling and Numbers Voice Database and FLEURS dataset
Creative Commons By Attribution 4.0, which applies to Multilingual Librispeech and Poly AI Minds 14
Proprietiary License of Munich AI Labs dataset
Public domain mark, which applies to PWR datasets
To use selected dataset, you also need to fill in the access forms on the specific datasets pages:
Common Voice: https://huggingface.co/datasets/mozilla-foundation/common_voice_13_0

Dataset Card for Polish ASR BIGOS corpora

Dataset Summary

The BIGOS (Benchmark Intended Grouping of Open Speech) corpora aims at simplifying the access and use of publicly available ASR speech datasets for Polish.
The initial release consist of test split with 1900 recordings and original transcriptions extracted from 10 publicly available datasets.

Supported Tasks and Leaderboards

The leaderboard with benchmark of publicly available ASR systems supporting Polish is under construction.
Evaluation results of 3 commercial and 5 freely available can be found in the paper.

Languages

Polish

Dataset Structure

Dataset consists audio recordings in WAV format and corresponding metadata.
Audio and metadata can be used in raw format (TSV) or via hugging face datasets library.

Data Instances

1900 audio files with original transcriptions are available in "test" split.
This consitutes 1.6% of the total available transcribed speech in 10 source datasets considered in the initial release.

Data Fields

Available fields:

file_id - file identifier
dataset_id - source dataset identifier
audio - binary representation of audio file
ref_original - original transcription of audio file
hyp_whisper_cloud - ASR hypothesis (output) from Whisper Cloud system
hyp_google_default - ASR hypothesis (output) from Google ASR system, default model
hyp_azure_default - ASR hypothesis (output) from Azure ASR system, default model
hyp_whisper_tiny - ASR hypothesis (output) from Whisper tiny model
hyp_whisper_base - ASR hypothesis (output) from Whisper base model
hyp_whisper_small - ASR hypothesis (output) from Whisper small model
hyp_whisper_medium - ASR hypothesis (output) from Whisper medium model
hyp_whisper_large - ASR hypothesis (output) from Whisper large (V2) model

Fields to be added in the next release:

ref_spoken - manual transcription in a spoken format (without normalization)
ref_written - manual transcription in a written format (with normalization)

Data Splits

Initial release contains only "test" split.
"Dev" and "train" splits will be added in the next release.

Dataset Creation

Curation Rationale

Polish ASR Speech Data Catalog was used to identify suitable datasets which can be repurposed and included in the BIGOS corpora.
The following mandatory criteria were considered:

Dataset must be downloadable.
The license must allow for free, noncommercial use.
Transcriptions must be available and align with the recordings.
The sampling rate of audio recordings must be at least 8 kHz.
Audio encoding using a minimum of 16 bits per sample.

Source Data

10 datasets that meet the criteria were chosen as sources for the BIGOS dataset.

The Common Voice dataset (mozilla-common-voice-19)
The Multilingual LibriSpeech (MLS) dataset (fair-mls-20)
The Clarin Studio Corpus (clarin-pjatk-studio-15)
The Clarin Mobile Corpus (clarin-pjatk-mobile-15)
The Jerzy Sas PWR datasets from Politechnika Wrocławska (pwr-viu-unk, pwr-shortwords-unk, pwr-maleset-unk). More info here
The Munich-AI Labs Speech corpus (mailabs-19)
The AZON Read and Spontaneous Speech Corpora (pwr-azon-spont-20, pwr-azon-read-20) More info here

Initial Data Collection and Normalization

Source text and audio files were extracted and encoded in a unified format.
Dataset-specific transcription norms are preserved, including punctuation and casing.
To strike a balance in the evaluation dataset and to facilitate the comparison of Word Error Rate (WER) scores across multiple datasets, 200 samples are randomly selected from each corpus.
The only exception is ’pwr-azon-spont-20’, which contains significantly longer recordings and utterances, therefore only 100 samples are selected.