lucataco/singing_voice_conversion

Amphion Singing Voice Conversion: DiffWaveNetSVC

Capabilities

No capability data available

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`source_audio` *	string (uri)	Input source audio file	`—`	—
`diffusion_inference_steps`	integer	Diffusion inference steps	`1000`	min: 0, max: 1000
`key_shift_mode`	integer	Key shift values	`0`	min: -6, max: 6
`pitch_shift_control`	string	Pitch shift control	`"Auto Shift"`	Auto Shift Key Shift
`target_singer`	string	Target singer to convert audio to	`"Taylor Swift"`	Adele Beyonce Bruno Mars John Mayer Michael Jackson Taylor Swift Jacky Cheung 张学友 Jian Li 李健 Feng Wang 汪峰 Faye Wong 王菲 Yijie Shi 石倚洁 Tsai Chin 蔡琴 Ying Na 那英 Eason Chan 陈奕迅 David Tao 陶喆

source_audio required string

Input source audio file

diffusion_inference_steps integer

Diffusion inference steps

Default: 1000 min: 0, max: 1000

key_shift_mode integer

Key shift values

Default: 0 min: -6, max: 6

pitch_shift_control string

Pitch shift control

Default: "Auto Shift"

Auto Shift Key Shift

target_singer string

Target singer to convert audio to

Default: "Taylor Swift"

Adele Beyonce Bruno Mars John Mayer Michael Jackson Taylor Swift Jacky Cheung 张学友 Jian Li 李健 Feng Wang 汪峰 Faye Wong 王菲 Yijie Shi 石倚洁 Tsai Chin 蔡琴 Ying Na 那英 Eason Chan 陈奕迅 David Tao 陶喆

Version: f29872ee3557 Updated: 2/26/2026 1.1K runs