Datasets:
Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
term: string
days: int64
records: int64
speakers: int64
chars: int64
words: int64
terms: string
to
{'terms': Value('string'), 'days': Value('int64'), 'records': Value('int64'), 'speakers': Value('int64'), 'chars': Value('int64'), 'words': Value('int64')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 478, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2818, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
term: string
days: int64
records: int64
speakers: int64
chars: int64
words: int64
terms: string
to
{'terms': Value('string'), 'days': Value('int64'), 'records': Value('int64'), 'speakers': Value('int64'), 'chars': Value('int64'), 'words': Value('int64')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Sejm Speeches Corpus — Kadencje VII, VIII, IX, X (2011-2027)
Korpus wypowiedzi z posiedzeń Sejmu RP, wygenerowany z oficjalnego API Sejmu (api.sejm.gov.pl).
Statystyki
| Metryka | Wartość |
|---|---|
| Kadencje | VII + VIII + IX + X |
| Dni posiedzeń | 825 |
| Rekordy | 157,714 |
| Mówcy | 2,588 |
| Znaki | 457,872,285 |
| Słowa | 63,884,505 |
| Tokeny (szac.) | ~83,049,856 |
| Rozmiar (ZIP) | 137 MB |
Plik
speeches_corpus_all.zip— skompresowany JSONL (157 714 rekordów)
Pola
| Pole | Typ | Opis |
|---|---|---|
text |
string | Treść wypowiedzi |
speaker |
string | Mówca (funkcja lub imię i nazwisko) |
date |
string | Data posiedzenia (ISO 8601) |
term |
string | Numer kadencji Sejmu (7, 8, 9, 10) |
source_url |
string | Link do materiałów źródłowych API Sejmu |
char_count |
int | Liczba znaków |
word_count |
int | Liczba słów |
has_events |
bool | Czy wypowiedzi towarzyszyły reakcje sali (oklaski itp.) |
Przetwarzanie
- Pobranie stenogramów z oficjalnego API Sejmu
- Parsowanie wypowiedzi z identyfikacją mówców
- Łączenie kolejnych wypowiedzi tego samego mówcy
- Filtrowanie wypowiedzi <50 słów
- Redakcja PII (emaile →
[EMAIL]) - Audyt jakości: PII, deduplikacja, dekontaminacja
- Merge per-kadencja + połączony
Audyt jakości
- PII: czyste (0 emaili, 0 telefonów, 0 PESEL)
- Duplikaty: 0
- Dekontaminacja: 0 nakładań (LLMzSzŁ, PoQuAD, PES)
- Werdykt: Obiecujący, gotowy do dalszej ewaluacji
Licencja
Materiały sejmowe są własnością publiczną (dokumenty urzędowe w rozumieniu art. 4 ustawy o prawie autorskim i prawach pokrewnych) — brak ograniczeń licencyjnych na wykorzystanie.
Pipeline
Kod pipeline: https://github.com/PiotrStyla/Sejm
Tryb tekstowy (domyślny): stenogramy → korpus tekstowy + audyt. Tryb audio (opcjonalny, modularny): + wideo HLS → audio → ASR → alignacja. Whisper jako baseline; architektura pozwala na podmianę na Cohere Transcribe / Qwen3-ASR.
- Downloads last month
- 81