Harmonic-Frontier-Audio/Overtone_Singing_Preview

这是一个高保真声乐数据集，展示了泛音歌唱——一种在世界多种歌唱传统中发现的复杂扩展声乐技巧，专为AI训练、音乐研究和创意声音开发设计。数据集包含高质量的录音，强调谐波隔离、频谱清晰度和受控的泛音运动，适用于AI语音建模、频谱分析、谐波合成和人类对齐的声乐控制研究。录音采用96 kHz / 24-bit的WAV格式，单声道，并包含详细的元数据。

A high-fidelity vocal dataset demonstrating overtone singing — a complex extended vocal technique found in multiple world singing traditions — designed for AI training, music research, and creative sound development. The recordings emphasize harmonic isolation, spectral clarity, and controlled overtone movement, making them valuable for AI voice modeling, spectral analysis, harmonic synthesis, and human-aligned vocal control research. Recorded at 96 kHz / 24-bit WAV format, mono, and includes detailed metadata.