ayousanz/beam-ja-yomi-bench

Beam-JA-Yomi Can Beam fix the misreadings of a Japanese text-to-speech front end? Japanese text-to-speech first turns kanji-and-kana text into a pronunciation (a katakana reading), usually with a dictionary-based converter such as OpenJTalk. The converter often gets numbers, units, symbols, words whose reading depends on context, and proper nouns wrong. This dataset is a small, human-approved test set for that step: each Japanese sentence comes with its reading as it is spoken… See the full description on the dataset page: https://huggingface.co/datasets/ayousanz/beam-ja-yomi-bench.

<iframe src="https://silytics.si/embed/item/df17ff86-7f59-4dbd-bb76-5c31b9435974" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.