We present mHuBERT-147, the first general-purpose massively multilingual
HuBERT speech representation model trained on 90K hours of clean, open-license
data. To scale up the multi-iteration HuBERT approach, we use faiss-based
clustering, achieving 5.2x faster label assignment over the original method. We
also apply a new multilingual batching up-sampling strategy, leveraging both
language and dataset diversity. After 3 training iterations and with only 95M
parameters, mHuBERT-147 outperforms larger models trained on substantially more
data. We rank second and first on the ML-SUPERB 10min/1h leaderboards
respectively, with SOTA scores for all LID tasks. Across ASR/LID tasks, our
model consistently surpasses XLS-R (300M params; 436K hours) and demonstrates
strong competitiveness against the much larger MMS (1B params; 491K hours). Our
findings suggest that mHuBERT-147 is a promising model for multilingual speech
processing tasks, offering an unprecedented balance between high performance
and parameter efficiency.

mHuBERT-147 是第一个以 90K 小时的干净、开源数据训练的通用多语言 HuBERT 语音表示模型，通过基于 faiss 的聚类实现了 5.2 倍更快的标签分配速度，同时应用了新的多语言批量上采样策略，在只有 95M 个参数的情况下，在经过 3 次训练迭代后能够胜过在更多数据上训练的更大模型，并且在 ML-SUPERB 10 分钟 / 1 小时排行榜上分别排名第二和第一，在所有语种识别任务中具有最高的性能得分。在 ASR/LID 任务中，我们的模型始终超越 XLS-R（300M 个参数；436K 小时）并且与更大的 MMS（1B 个参数；491K 小时）具有很强的竞争力。我们的研究结果表明，mHuBERT-147 是一个有前景的多语言语音处理模型，提供了高性能和参数效率之间的前所未有的平衡。