Common Phone

Common Phone是一个多语种的语音数据集，由Mozilla的Common Voice项目中的超过11,000名贡献者录制，总时长约116小时。该数据集特别强调性别平衡，并包含自动生成的音素分割。数据集内容包括英语、德语、西班牙语、法语、意大利语和俄语六种语言，每种语言都有详细的年龄和性别分布。创建过程中，通过自动化的音素标注程序确保了音素标注的准确性。Common Phone数据集主要用于支持声学模型的研究与应用，特别是在多语言环境下提高模型的泛化能力和鲁棒性。

Common Phone is a multilingual speech dataset recorded by more than 11,000 contributors from Mozilla's Common Voice project, with a total duration of approximately 116 hours. It places special emphasis on gender balance and incorporates automatically generated phoneme segmentation. The dataset covers six languages, namely English, German, Spanish, French, Italian and Russian, with detailed age and gender distributions for each language. During its creation, an automated phoneme annotation program was adopted to ensure the accuracy of phoneme annotations. The Common Phone dataset is primarily used to support research and applications of acoustic models, especially to improve the generalization ability and robustness of models in multilingual environments.