600小时菲律宾英语分轨自然对话手机采集语音数据

菲律宾英语分轨自然对话手机采集语音数据，基于常见主题来模拟录制。此数据集标注了文本内容、句时间戳、说话人身份、性别等多种属性，由来自不同地域和文化背景的菲律宾本土人录制，准确性高，为语音识别相关研究及应用提供了丰富的资源，经多家AI公司验证：有助于模型面对真实世界的多样性时能够表现出色。我们严格遵循数据保护法规和隐私规定，确保数据采集、存储和使用的过程中维护用户的隐私和合法权益，所有数据均遵循GDPR, CCPA, PIPL。This is a mobile-collected naturally conversational Philippine English speech dataset with separate speaker audio tracks, recorded based on common daily conversational themes to simulate real-world dialogues. This dataset is annotated with multiple comprehensive attributes including speech transcripts, sentence-level timestamps, speaker identities, and genders. It is recorded by native Filipino speakers from diverse regional and cultural backgrounds, ensuring high data accuracy. It provides abundant resources for speech recognition-related research and applications. Verified by several AI companies, this dataset helps models deliver excellent performance when facing real-world diversity. We strictly comply with data protection regulations and privacy rules, ensuring that user privacy and legitimate rights and interests are safeguarded throughout the processes of data collection, storage, and utilization. All data in this dataset adheres to GDPR, CCPA, and PIPL.

搜集汇总

背景与挑战

背景概述

该数据集包含600小时的菲律宾英语自然对话语音数据，通过手机采集并分轨处理，由约800名菲律宾本土人在安静室内环境中录制，话题多样以确保对话流畅自然。数据标注了文本内容、时间戳、说话人身份和性别等属性，词准确率达98%，适用于语音识别和声纹识别研究，并严格遵循数据隐私法规。