YuFeng926 OpenTune[FREE]@基于神经声码器的开源AI人声音高修正工具（Win+Mac）

简介:
OpenTune v1.8.1独立VST3 WiN MAC [免费]
免费 | Windows Mac | 563 MB
OpenTune-人工智能音高校正软件
保留共振峰，无失真音高变换
OpenTune是一款基于神经声码器的开源音高校正工具。与传统的直接移位音高的DSP算法不同，它使用nsf-hifigan声码器来重新生成人声，同时保留原始共振峰，即使进行极端的音高调整也能保持自然，稳定的音质。
该编辑器以图形化调优为核心，而非采用一键式自动校正。您可以绘制音高曲线、编辑检测到的注释、使用定位工具、自动检测关键点、校正到常用比例，以及在主机集成行为正常时在ARA2-capable主机中工作。
主要特点
每注EQ处理:
便捷且精细的编辑功能，这是其他DAW插件所无法提供的。
混合音高变换：
通过与DSP算法相结合，它可以在一个八度音阶内实现更自然的换位-用于小校正的第一层声音质量，以及用于大移位的行业领先的自然结果。
保留的共振峰:
改变音高几乎不会影响音色-您可以立即听到的差异
人工智能重合成：
基于深度学习的声码器，而非传统的音高变换
自动调谐/类似Melodyne的工作流程:
手绘、笔记和锚定工具-快速进入流程所需的一切
内置自动音键检测：
立即开始校正-即使没有校正也准确，甚至更准确
多尺度校正:
色度，大调，小调，五声，多利安，混合音，谐波小调
音频格式:
导入WAV、FLAC、OGG、MP3
人工智能引擎：
FCPE音高提取pc-nsf HifiGAN声码器 (ONNX运行时推断)
双格式输出：
独立可执行VST3插件 (支持ARA2扩展)
GPU加速推理：
Windows DirectML / macOS CoreML
多语言界面：
中文/俄语/日语/英语/西班牙语
开源且免费：
永久免费，社区驱动，持续迭代

简介:
OpenTune v1.8.1 STANDALONE VST3 WiN MAC [FREE]
FREE | WiN MAC | 563 MB
OpenTune – AI Pitch Correction Software
Preserving formants, pitch shifting without distortion
OpenTune is an open-source pitch correction tool based on neural vocoders. Unlike traditional DSP algorithms that directly shift pitch, it uses the NSF-HiFiGAN vocoder to regenerate vocals while preserving the original formants, maintaining natural, solid sound quality even with extreme pitch adjustments.
The editor is built around graphical tuning rather than one-knob automatic correction. You can draw pitch curves, edit detected notes, use anchor tools, detect key automatically, correct to common scales, and work in ARA2-capable hosts when the host integration behaves correctly.
Key Features
Per-note EQ processing:
convenient, fine-grained editing that other DAW plugins can’t deliver
Hybrid pitch shifting:
By combining with DSP algorithms, it achieves more natural transposition within one octave — first-tier sound quality for small corrections, and industry-leading natural results for large shifts.
Formants preserved:
Changing pitch barely affects the timbre — a difference you can hear instantly
AI resynthesis:
Deep learning-based vocoder, not traditional pitch shifting
Auto-Tune/Melodyne-like workflow:
Hand-drawing, note, and anchor tools — everything you need to get into the flow fast
Built-in automatic key detection:
Start correcting right away — accurate even without correction, and even more accurate after
Multi-scale correction:
Chromatic, Major, Minor, Pentatonic, Dorian, Mixolydian, Harmonic Minor
Audio formats:
WAV, FLAC, OGG, MP3 import
AI engine:
FCPE pitch extraction + PC-NSF HifiGAN vocoder (ONNX Runtime inference)
Dual-format output:
Standalone executable + VST3 plugin (with ARA2 extension support)
GPU-accelerated inference:
Windows DirectML / macOS CoreML
Multi-language interface:
Chinese / English / Japanese / Russian / Spanish
Open source and free:
Permanently free, community-driven, continuously iterating