Preprint
Dialect and Gender Bias in YouTube's Spanish Captioning System
Iris Dania Jimenez and Christoph Kern, arXiv (Computation and Language, cs.CL), preprint 2602.24002, 2026
Read the originalRead on October 17, 2026
- What it is
- A 2026 preprint that measures the automatic captions YouTube makes for Spanish speakers from seven Latin American dialects, comparing word error rate by dialect and by gender. It reports that Argentinian speech had the highest error rate and that female Puerto Rican speech had the lowest.
- How much weight it carries
- A preprint that has not been peer reviewed. It tests one captioning system (YouTube's automatic captions, not dictation) on recordings from different datasets for each dialect, so the dialect comparison is indicative rather than a controlled test. It covers Latin American dialects only, not Castilian Spanish.
Facts taken from this source
- In a 2026 test of YouTube's automatic Spanish captions, Argentinian Spanish had the highest overall word error rate of the dialects tested, at 24 percent.
“Argentinian Spanish yields the highest overall WER at 24%” (Results (Table 3 discussion), PDF body)
- The authors found systematic differences in how accurately YouTube's Spanish captions handled different dialects, which they attribute to specific dialects rather than to chance.
“we identify systematic disparities which can be attributed to specific dialects” (Abstract)
- In a 2026 test of YouTube's automatic Spanish captions, Mexican Spanish had an overall word error rate of 18 percent.
“Mexican Spanish exhibit an overall WER value of 18%” (Results (Table 3 discussion), PDF body)