Datasets:
sentence_id string | origin string | eng string | translations list |
|---|---|---|---|
tatoeba-flatten-00000000 | tatoeba-flatten | $8.50, please. | [
{
"lang": "deu",
"text": "8,50 $, bitte."
},
{
"lang": "spa",
"text": "$8,50, por favor."
},
{
"lang": "kor",
"text": "8.50달러 주세요."
},
{
"lang": "nld",
"text": "$8,50, alstublieft."
},
{
"lang": "por",
"text": "US$ 8,50, por favor."
},
{
"lang": "jpn",... |
tatoeba-flatten-00000001 | tatoeba-flatten | '-osity' is an abstract noun word ending created from the ending of an '-ous' adjective. | [
{
"lang": "deu",
"text": "„-osity“ ist eine abstrakte Nomenwortendung, die aus der Endung eines Adjektivs auf „-ous“ gebildet wird."
},
{
"lang": "spa",
"text": "'-osity' es una terminación de palabra sustantiva abstracta creada a partir de la terminación de un adjetivo '-ous'."
},
{
"la... |
tatoeba-flatten-00000002 | tatoeba-flatten | 'A' comes before 'B' in the alphabet. | [
{
"lang": "deu",
"text": "Im Alphabet kommt „A“ vor „B“."
},
{
"lang": "spa",
"text": "La 'A' va antes de la 'B' en el alfabeto."
},
{
"lang": "kor",
"text": "알파벳에서는 'A'가 'B' 앞에 옵니다."
},
{
"lang": "nld",
"text": "'A' komt voor 'B' in het alfabet."
},
{
"lang": "po... |
tatoeba-flatten-00000003 | tatoeba-flatten | 'Aha', they will say. | [
{
"lang": "deu",
"text": "Aha, werden sie sagen."
},
{
"lang": "spa",
"text": "'Ajá', dirán."
},
{
"lang": "kor",
"text": "'아하'라고 그들은 말할 것입니다."
},
{
"lang": "nld",
"text": "'Aha', zullen ze zeggen."
},
{
"lang": "por",
"text": "'Ah', dirão."
},
{
"lang... |
tatoeba-flatten-00000004 | tatoeba-flatten | 'Biology' is a Greek word that means 'the study of living organisms'. | [
{
"lang": "deu",
"text": "„Biologie“ ist ein griechisches Wort und bedeutet „Studium lebender Organismen“."
},
{
"lang": "spa",
"text": "\"Biología\" es una palabra griega que significa \"el estudio de los organismos vivos\"."
},
{
"lang": "kor",
"text": "생물학(Biology)은 그리스어로 '살아있는 유기... |
tatoeba-flatten-00000005 | tatoeba-flatten | 'Call off your dogs, or they will kill me. What harm have I ever done to you?' At these words, coming from a bear, the prince was so startled that for a moment he stood stock-still. | [
{
"lang": "deu",
"text": "„Ruf deine Hunde zurück, sonst beißen sie mich tot! Was habe ich dir je zuleid getan?“ Auf diese Worte hin, die von einem Bären kamen, war der Prinz für einen Moment so erschrocken, dass er wie versteinert stand."
},
{
"lang": "spa",
"text": "'Llama a tus perros o me ma... |
tatoeba-flatten-00000006 | tatoeba-flatten | 'Cause I don't really change my E-mail address. | [
{
"lang": "deu",
"text": "Weil ich meine E-Mail-Adresse nicht wirklich ändere."
},
{
"lang": "spa",
"text": "Porque realmente no cambio mi dirección de correo electrónico."
},
{
"lang": "kor",
"text": "왜냐면 난 내 이메일 주소를 별로 바꾸지 않거든."
},
{
"lang": "nld",
"text": "Omdat ik mij... |
tatoeba-flatten-00000007 | tatoeba-flatten | 'Childhood friends' are scary. | [
{
"lang": "deu",
"text": "„Freunde aus der Kindheit“ sind gruselig."
},
{
"lang": "spa",
"text": "Los 'amigos de la infancia' dan miedo."
},
{
"lang": "kor",
"text": "'소꿉친구'는 무섭다."
},
{
"lang": "nld",
"text": "'Jeugdvrienden' zijn eng."
},
{
"lang": "por",
"te... |
tatoeba-flatten-00000008 | tatoeba-flatten | 'Cromulent' is a perfectly cromulent word. | [
{
"lang": "deu",
"text": "„Cromulent“ ist ein vollkommen cromulentes Wort."
},
{
"lang": "spa",
"text": "\"Cromulento\" es una palabra perfectamente cromulente."
},
{
"lang": "kor",
"text": "'Cromulent'는 완벽하게 cromulent 단어입니다."
},
{
"lang": "nld",
"text": "'Cromulent' is e... |
tatoeba-flatten-00000009 | tatoeba-flatten | 'Cuz it's a giant pain in the arse, if I don't wanna do something I just won't do it. | [
{
"lang": "deu",
"text": "Weil es ein riesiger Krampf ist. Wenn ich etwas nicht tun will, dann tue ich es einfach nicht."
},
{
"lang": "spa",
"text": "Porque es un dolor de cabeza gigante, si no quiero hacer algo, simplemente no lo haré."
},
{
"lang": "kor",
"text": "왜냐면 그건 정말 골치 아픈 ... |
tatoeba-flatten-00000010 | tatoeba-flatten | 'Don't worry. I'm doing fine. :-) ' "Eh ... smiley-face?" | [
{
"lang": "deu",
"text": "„Keine Sorge. Mir geht’s gut. :-)“ „Äh … Smiley?“"
},
{
"lang": "spa",
"text": "'No te preocupes. Estoy bien. :-) ' \"Eh... ¿cara sonriente?\""
},
{
"lang": "kor",
"text": "'괜찮아요. 나는 잘 지내고 있다. :-) ' \"어... 웃는 얼굴?\""
},
{
"lang": "nld",
"text": "'... |
tatoeba-flatten-00000011 | tatoeba-flatten | 'E' is not a Spanish word. "Yes, it is. You're mistaken." | [
{
"lang": "deu",
"text": "„E“ ist kein spanisches Wort. „Doch, das ist es. Sie irren sich.“"
},
{
"lang": "spa",
"text": "'E' no es una palabra española. \"Sí, lo es. Estás equivocado.\""
},
{
"lang": "kor",
"text": "'E'는 스페인어가 아닙니다. \"그렇습니다. 당신은 착각하고 있습니다.\""
},
{
"lang": "n... |
tatoeba-flatten-00000012 | tatoeba-flatten | 'Four' is an unlucky number in Japan. | [
{
"lang": "deu",
"text": "In Japan ist die Zahl „Vier“ eine Unglückszahl."
},
{
"lang": "spa",
"text": "El 'cuatro' es un número de mala suerte en Japón."
},
{
"lang": "kor",
"text": "'4'는 일본에서 불길한 숫자입니다."
},
{
"lang": "nld",
"text": "‘Vier’ is in Japan een ongeluksgetal.... |
tatoeba-flatten-00000013 | tatoeba-flatten | 'Fox's wedding'. "Er ...?" "Today, with an 81 percent likelihood, it will be clear skies but, depending on location, there will be sun showers." | [
{
"lang": "deu",
"text": "'Hochzeit des Fuchses'. \"Äh...?\" \"Heute ist mit 81-prozentiger Wahrscheinlichkeit klarer Himmel zu erwarten, je nach Standort gibt es aber auch mal sonnige Schauer.\""
},
{
"lang": "spa",
"text": "'La boda de Fox'. \"¿Eh...?\" \"Hoy, con una probabilidad del 81 por c... |
tatoeba-flatten-00000014 | tatoeba-flatten | 'Guillaume and Lebas' will that not make a good business name? We might add, 'and Co.' to round off the firm's signature. | [
{
"lang": "deu",
"text": "Ist „Guillaume und Lebas“ nicht ein guter Firmenname? Wir könnten „und Co.“ hinzufügen, um die Unterschrift der Firma abzurunden."
},
{
"lang": "spa",
"text": "'Guillaume y Lebas' ¿no sería un buen nombre comercial? Podríamos agregar \"y compañía\". para redondear la fi... |
tatoeba-flatten-00000015 | tatoeba-flatten | 'Hayastan' is Armenia's name in Armenian. | [
{
"lang": "deu",
"text": "„Hayastan“ ist Armeniens Name auf Armenisch."
},
{
"lang": "spa",
"text": "'Hayastan' es el nombre de Armenia en armenio."
},
{
"lang": "kor",
"text": "'하야스탄'은 아르메니아어로 아르메니아의 이름이다."
},
{
"lang": "nld",
"text": "'Hayastan' is de naam van Armenië i... |
tatoeba-flatten-00000016 | tatoeba-flatten | 'Hinder me? Thou fool. No living man may hinder me!' Then Merry heard of all sounds in that hour the strangest. It seemed that Dernhelm laughed, and the clear voice was like the ring of steel. 'But no living man am I! You look upon a woman. Éowyn I am, Éomund's daughter.' | [
{
"lang": "deu",
"text": "„Mich behindern? Du Narr. Kein lebender Mensch darf mich behindern!“ Dann hörte Merry von allen Geräuschen in dieser Stunde das seltsamste. Es schien, als würde Dernhelm lachen, und die klare Stimme klang wie der Klang von Stahl. „Aber ich bin kein lebender Mensch! Du siehst eine F... |
tatoeba-flatten-00000017 | tatoeba-flatten | 'Is that vodka?' Margarita asked weakly. The cat jumped up from its chair in indignation. 'Excuse me, your majesty,' he squeaked, 'do you think I would give vodka to a lady? That is pure spirit!' | [
{
"lang": "deu",
"text": "„Ist das Wodka?“, fragte Margarita schwach. Die Katze sprang empört von ihrem Stuhl auf. „Entschuldigen Sie, Eure Majestät“, quiekte sie, „glauben Sie, ich würde einer Dame Wodka geben? Das ist purer Schnaps!“"
},
{
"lang": "spa",
"text": "'¿Eso es vodka?' Margarita pre... |
tatoeba-flatten-00000018 | tatoeba-flatten | 'It must be very scary to be you,' Janina said in a whisper. | [
{
"lang": "deu",
"text": "„Es muss sehr beängstigend sein, du zu sein“, sagte Janina flüsternd."
},
{
"lang": "spa",
"text": "\"Ser tú debe dar mucho miedo\", dijo Janina en un susurro."
},
{
"lang": "kor",
"text": "'당신이 된다는 건 정말 무서운 일이겠군요.' Janina가 속삭이듯 말했습니다."
},
{
"lang": ... |
tatoeba-flatten-00000019 | tatoeba-flatten | 'Kiss my nose', I answered as I didn't want to say 'ass' to her. | [
{
"lang": "deu",
"text": "„Küss meine Nase“, antwortete ich, da ich nicht „Arsch“ zu ihr sagen wollte."
},
{
"lang": "spa",
"text": "'Bésame la nariz', le respondí porque no quería decirle 'culo'."
},
{
"lang": "kor",
"text": "'Kiss my nose', 나는 그녀에게 '엉덩이'라는 말을 하고 싶지 않아 대답했다."
},
... |
tatoeba-flatten-00000020 | tatoeba-flatten | 'La La Land' was mistakenly announced as the winner. | [
{
"lang": "deu",
"text": "„La La Land“ wurde fälschlicherweise zum Gewinner erklärt."
},
{
"lang": "spa",
"text": "'La La Land' fue anunciada por error como ganadora."
},
{
"lang": "kor",
"text": "'라라랜드'가 우승자로 잘못 발표됐다."
},
{
"lang": "nld",
"text": "'La La Land' werd per o... |
tatoeba-flatten-00000021 | tatoeba-flatten | 'Madame Butterfly' is set in Nagasaki. | [
{
"lang": "deu",
"text": "„Madame Butterfly“ spielt in Nagasaki."
},
{
"lang": "spa",
"text": "'Madame Butterfly' está ambientada en Nagasaki."
},
{
"lang": "kor",
"text": "'나비부인'은 나가사키를 배경으로 한다."
},
{
"lang": "nld",
"text": "'Madame Butterfly' speelt zich af in Nagasaki.... |
tatoeba-flatten-00000022 | tatoeba-flatten | 'Managing' simply means taking the blame for everyone's mistakes. | [
{
"lang": "deu",
"text": "„Managen“ bedeutet einfach, die Schuld für die Fehler aller auf sich zu nehmen."
},
{
"lang": "spa",
"text": "\"Gestionar\" significa simplemente asumir la culpa de los errores de todos."
},
{
"lang": "kor",
"text": "'관리'란 단순히 모든 사람의 실수에 대해 책임을 지는 것을 의미합니다."... |
End of preview. Expand in Data Studio
natgillin/n-way-translations
Multilingual N-way parallel corpus: every row is one logical sentence expressed in as many languages as the upstream source has it.
Schema
| column | type | description |
|---|---|---|
sentence_id |
string | <origin>-<NNNNNNNN> — stable within an origin |
origin |
string | upstream identifier (e.g. tatoeba-flatten) |
eng |
string | English text (also kept inside translations); null if the row has no English |
translations |
list<struct<lang: string, text: string>> | one entry per non-null language: [{lang: "eng", text: "..."}, {lang: "deu", text: "..."}, ...] |
eng is duplicated as a top-level column for easy filtering / sorting /
joining; the same value is also kept inside translations under
lang == "eng" so each row stays self-contained.
translations uses list<struct<lang,text>> (not map<string,string>)
because the HF dataset viewer does not yet render map columns. Convert
back to a Python dict in one line: dict((e['lang'], e['text']) for e in row['translations']).
Configs / layout
One config per upstream origin.
data/
tatoeba-flatten/
train-00000-of-NNNNN.parquet
...
Origins
| origin | rows | source | languages |
|---|---|---|---|
tatoeba-flatten |
215,912 | alvations/tatoeba-flatten |
up to 50 (ISO 639-3): eng, deu, spa, kor, nld, por, jpn, ita, pol, fra, ara, ukr, rus, heb, hun, ell, bul, swe, hrv, tur, vie, ben, srp, slk, fin, ces, bel, hin, ron, dan, nob, uzb, isl, ind, cat, aze, msa, tha, urd, pan, est, mar, tam, guj, slv, mal, tel, kan, ori, zho |
Loading
from datasets import load_dataset
ds = load_dataset("natgillin/n-way-translations", "tatoeba-flatten", split="train")
ex = ds[0]
print(ex["eng"]) # English text
print({e["lang"]: e["text"] for e in ex["translations"]}) # full dict view
License
Inherits per-corpus terms from the upstream sources. Tatoeba: https://tatoeba.org.
- Downloads last month
- 28