sarvamai/tts-robustness-benchmark
收藏Hugging Face2026-02-06 更新2026-02-07 收录
下载链接：
https://hf-mirror.com/datasets/sarvamai/tts-robustness-benchmark
下载链接
链接失效反馈官方服务：
资源简介：
TTS鲁棒性基准测试是一个用于评估文本到语音（TTS）系统在不同领域中的字符错误率（CER）的样本集合。该基准测试确保每个句子-语言对仅出现一次，从而提供干净可靠的TTS鲁棒性度量。数据集覆盖11种语言（如印地语、泰米尔语、卡纳达语等）和7个关键压力测试类别（如数字、STEM、印度命名实体等），总样本数为959。
The TTS Robustness Benchmark is a set of evaluation samples used to calculate domain-wise CER (Character Error Rate) scores across 7 critical stress-test categories. This benchmark ensures that each sentence-language pair appears exactly once, providing a clean and reliable metric for TTS robustness. The dataset covers 11 languages (e.g., Hindi, Tamil, Kannada) and 7 core domains (e.g., Numerics, STEM, Indian Named Entities), with a total of 959 samples.
提供机构：
sarvamai
