Yali语音：带完整声调的孤立普通话音节录音库

这些音节由程雅丽（Cheng Ya Li，汉字待确认）于2008年6月为Gradint程序录制。音频编辑由黄冠能（Cameron Wong，Ekho语音合成器的开发者）和Silas S. Brown（赛乐思）使用Audacity和Praat完成。第一声和第六声：已进行轻微音调调整以补偿录音时段间的音调漂移. 第二、三、四、五声：保留从语境中提取的原始音调. 第三声：录音为自然语音中常用的"半三声". 第六声：通常在第三声后使用的高音中性调

These syllables were recorded in June 2008 by Cheng Ya Li (Chinese characters to be confirmed) for the Gradint program. Audio editing was performed by Cameron Wong, developer of the Ekho speech synthesizer, and Silas S. Brown (also known as Sailosi) using the audio editing tools Audacity and Praat.
For the first and sixth tones: Minor pitch adjustments were made to compensate for pitch drift across different recording sessions.
For the second, third, fourth and fifth tones: The original pitch contours extracted from their respective contexts were retained.
For the third tone: The recordings adopted the "half third tone", a common prosodic variant in natural speech.
For the sixth tone: A high-pitched neutral tone typically used after a third tone.

搜集汇总

背景与挑战

背景概述

该数据集是一个包含完整声调覆盖的汉语孤立音节录音库，由程雅丽于2008年6月为Gradint程序录制，音频经过编辑处理以优化音调一致性。录音库提供了技术说明，包括对第一声和第六声的音调调整，以及第二至五声的原始音调保留，适用于语音合成研究。