海天智源

该语音数据集包含100多种语言及方言，涵盖中文，中文方言，亚洲、欧洲、美洲、中东及非洲等主要国家及地区语言/口音，累计约10万小时语音时长。数据集主要应用于人工智能领域相关的软件及技术（语言识别方向），其应用场景有IT互联网，智能搜索，自动驾驶，智能语音，机器人，智慧学习，智慧金融等。

This speech dataset comprises over 100 languages and dialects, covering Mandarin, Chinese dialects, as well as languages and accents from major countries and regions across Asia, Europe, the Americas, the Middle East and Africa, with a total accumulated speech duration of approximately 100,000 hours. It is primarily applied to AI-related software and technologies targeting speech recognition. Its application scenarios include IT and internet, intelligent search, autonomous driving, intelligent voice, robotics, smart learning, and smart finance, among others.

搜集汇总

背景与挑战

背景概述

海天智源是一个大规模的语音数据集，涵盖100多种语言及方言，包括中文、中文方言以及全球主要地区的语言，累计语音时长约10万小时。该数据集专为人工智能语音识别技术设计，适用于IT互联网、自动驾驶、智能语音、机器人等多个领域的应用开发。