Skip to content
SPECIFICTION
Get listed

SoulX-Singer

github.com

Free plan WebLinuxSelf-hosted
Visit github.com
Spec score 8.1 #9 of 55 in AI song cover generators

What it is

SoulX-Singer is a singing voice synthesis model that generates voices for singers it has not been tuned on. Pitch, rhythm, and expression can be guided by melody-conditioned F0 contours or score-conditioned MIDI notes. Its voice-conversion system changes raw singing audio to a target singer’s voice while retaining melody, rhythm, and lyrics, without requiring lyric or MIDI transcriptions. It supports Mandarin Chinese, English, and Cantonese, plus singer-timbre cloning, cross-lingual synthesis, and lyric editing while retaining natural prosody. Preprocessing tools handle vocal separation, dereverberation, F0 extraction, voice activity detection, and lyric and note transcription. Users can edit MIDI metadata, including lyrics, phoneme alignment, pitches, and durations, before synthesis. Soul-AILab provides generation and conversion demos and a separate MIDI editor through Hugging Face Spaces; the repository also supports local inference. Code and model weights use the Apache 2.0 license. Maintainers caution that automatic preprocessing may misalign singing with lyrics and notes, so manual correction may be needed.

Spec sheet

Purpose
SoulX-Singer is a high-fidelity zero-shot singing voice synthesis model for generating realistic voices for unseen singers.Read 1 Oct 2026 · github.com
Deployment
The repository can be cloned, installed with Conda and pip, and run locally through Python web UI scripts.Read 1 Oct 2026 · github.com
Support
The project lists three contact emails and invites technical discussion through WeChat or Soul app groups.Read 1 Oct 2026 · github.com
License
The code and model weights are released under the Apache 2.0 license for researchers and developers to use.Read 1 Oct 2026 · github.com
Control modes
It supports melody-conditioned F0 contour control and score-conditioned MIDI note control for pitch, rhythm, and expression.Read 1 Oct 2026 · github.com
Core function
SoulX-Singer is a high-fidelity zero-shot singing voice synthesis model for generating realistic voices for unseen singers.Read 1 Oct 2026 · github.com
Dataset scale
The stated training dataset contains more than 42,000 hours of aligned vocals, lyrics, and notes.Read 1 Oct 2026 · github.com
Editing and cloning
Features include singer-timbre cloning, cross-lingual synthesis, and lyric editing while preserving natural prosody.Read 1 Oct 2026 · github.com
Languages
The system supports Mandarin Chinese, English, and Cantonese.Read 1 Oct 2026 · github.com
License and safety
The project uses Apache 2.0 and asks users to respect intellectual property, privacy, and consent and avoid unauthorized impersonation or deceptive audio.Read 1 Oct 2026 · github.com
Local deployment
The repository supports local inference through Conda with Python 3.10, pip-installed dependencies, and local WebUI scripts.Read 1 Oct 2026 · github.com
MIDI editing
A MIDI Editor supports editing lyrics, phoneme alignment, note pitches, and durations before inference.Read 1 Oct 2026 · github.com
MIDI integration
Generated metadata can be exported to MIDI, edited for lyrics, phoneme alignment, pitches, and durations, and imported back for synthesis.Read 1 Oct 2026 · github.com
Model distribution
Pretrained synthesis, conversion, and preprocessing models are downloaded through Hugging Face Hub commands.Read 1 Oct 2026 · github.com
Notable limitation
The maintainers warn that automatic preprocessing may misalign singing audio with lyrics and notes and recommend manual correction.Read 1 Oct 2026 · github.com
Online access
Soul-AILab provides a running SoulX-Singer demo on Hugging Face Spaces and a separate running MIDI Editor Space.Read 1 Oct 2026 · huggingface.co
Pitch and score control
It supports melody-conditioned F0-contour control and score-conditioned MIDI-note control for pitch, rhythm, and expression.Read 1 Oct 2026 · github.com
Preprocessing
Its preprocessing toolkit performs vocal separation and dereverberation, F0 extraction, voice activity detection, lyrics transcription, and note transcription.Read 1 Oct 2026 · github.com
Training data
The project reports more than 42,000 hours of aligned vocal, lyric, and note data.Read 1 Oct 2026 · arxiv.org
Usage restrictions
The maker asks users to respect intellectual property, privacy, and consent and prohibits unauthorized impersonation or deceptive audio.Read 1 Oct 2026 · github.com
Voice conversion
SoulX-Singer-SVC converts raw singing audio into a target singer’s voice while preserving melody, rhythm, and lyrics without lyric or MIDI transcriptions.Read 1 Oct 2026 · github.com
Web access
The maker provides a SoulX-Singer singing-generation and vocal-conversion demo on Hugging Face Spaces.Read 1 Oct 2026 · huggingface.co
Zero-shot operation
The model generates voices for unseen singers without fine-tuning or per-speaker fine-tuning.Read 1 Oct 2026 · github.com

Plans and prices

SoulX-Singer Free Researchers and developers are free to use the code and model weights · Apache 2.0 license Read 1 Oct 2026 · github.com

Every SoulX-Singer plan and price →

Strong points and limits

Strong points

  • Generates voices for unseen singers without fine-tuning
  • Supports Mandarin, English, and Cantonese
  • Provides MIDI editing before synthesis
  • Includes generation, conversion, and MIDI editor demos
  • Apache 2.0 code and weights

Limits

  • Automatic preprocessing may misalign audio, lyrics, and notes
  • Local inference requires Conda, Python 3.10, and dependencies

Who it suits

SoulX-Singer suits researchers and developers seeking a locally deployable or demo-accessible singing synthesis and voice-conversion model. Its license permits use of the code and weights, while the maintainers ask users to respect intellectual property, privacy, and consent.

Verdict

SoulX-Singer offers voice synthesis and conversion with melody and MIDI controls, alongside online demos and local inference. Plan for manual correction when automatic preprocessing misaligns the singing and its lyrics or notes.

Where Specifiction places it

Alternatives to SoulX-SingerAll alternatives →