appautomaton/step-audio-editx-8bit-mlx
Step-Audio-EditX — MLX 8-bit
This repository contains a self-contained pure-MLX int8 conversion of Step-Audio-EditX for local voice cloning and expressive audio editing on Apple Silicon. All pipeline components are stored as .safetensors — no PyTorch, ONNX, or NumPy files are required at inference time.
Model Details
- Developed by: AppAutomaton
- Upstream model: `stepfun-ai/Step-Audio-EditX`
- Task: zero-shot voice cloning, expressive audio editing
- Runtime: MLX on Apple Silicon
- Precision: int8 for Step1 LM, Flow model, and VQ02 tokenizer; bf16 for the rest
- Total size: ~4.1 GB (down from ~7.7 GB upstream)
Bundle Contents
This bundle is self-contained — all weights are packaged in one repository.
How to Get Started
Download the bundle:
hf download appautomaton/step-audio-editx-8bit-mlx \
--local-dir models/stepfun/step_audio_editx/mlx-int8Voice cloning:
python scripts/generate/step_audio_editx.py \
--prompt-audio reference.wav \
--prompt-text "Transcript of reference audio." \
-o cloned.wav \
clone --target-text "New speech in the cloned voice."Audio editing (change emotion):
python scripts/generate/step_audio_editx.py \
--prompt-audio input.wav \
--prompt-text "Transcript of input audio." \
-o happy.wav \
edit --edit-type emotion --edit-info happySupported Edit Types
Architecture
Five-stage pipeline, all running pure MLX with bf16 activations:
- Step1 LM (3.5B params, int8) — autoregressive dual-codebook token generation
- CampPlus (bf16) — speaker embedding extraction from reference audio
- Flow conditioner (bf16) — conditions generation on speaker embedding
- Flow model (int8) — flow-matching mel spectrogram generation
- HiFT vocoder (bf16) — mel spectrogram to waveform
The VQ02 and VQ06 tokenizers encode reference audio into dual codebook tokens consumed by Step1.
Performance
On Apple Silicon with int8 weights and bf16 activations, real-time factor (RTF) is approximately 1.46x for voice cloning — faster than real-time.
Links
- Source code: `mlx-speech`
- Project page: appautomaton.com/mlx-speech
- Upstream model: `stepfun-ai/Step-Audio-EditX`
- Technical report: arXiv:2511.03601
- More examples: AppAutomaton
License
Apache 2.0 — following the upstream license published with `stepfun-ai/Step-Audio-EditX`.
