Spanish TN Corpus

This dataset is given the text, perform the annotation and transcription of non-textual symbols, numbers, and letters.
Specifications:
ID:
King-NLP-189
Size:
116806 sentences
Language:
Spanish
Quantity
116,357 entries
Country
Spain
Language Code
ES-ES
Accuracy Rate
Label Accuracy 95%

People also searched for

High-quality competition question bank dataset
Original questions created by experts and scholars, covering mathematics, physics, chemistry, biology, and informatics. They have passed the initial screening. The reasoning is relatively rigorous and the provided steps can be verified. They focus on the core content from senior undergraduate to graduate level, and can effectively test the upper limit of the model's reasoning. The questions are mainly in Chinese, with a few purely English ones, and standard LaTeX symbols are embedded.
Danish multiple round conversation Text dataset
A complete set of multi-round conversations involving two or more people, covering the fields of finance, retail and travel.
Chinese-English Parallel Corpus
Parallel corpora
English-Chinese Parallel Corpus
Parallel corpora

Join our newsletter to stay updated

Thank you for signing up!

Stay informed and ahead with the latest updates, insights, and exclusive content delivered straight to your inbox.

By subscribing you agree to with our Privacy Policy and provide consent to receive updates from our company.