Mixed-precision
CoreML export of Supertonic-3's four non-autoregressive flow-matching graphs,
tuned for
Apple Neural Engine
residency on iOS 18+ / macOS 15+. The precision split was chosen by
measured end-to-end SNR
— flow-matching is an ODE and its trajectory is precision-sensitive, so the
graphs that drive it stay FP32:
TextEncoder
+
VectorEstimator
→
FP32
(drive the ODE trajectory; the vector-estimator is
ANE-ineligible regardless of precision).
Vocoder
+
DurationPredictor
→
FP16
(the vocoder is single-shot and the 100%-ANE win).
Controlled shared-noise A/B vs the FP32 reference:
mag-STFT SNR 47–51 dB
(en/de/ko), max|Δ|
6e-3 —
transparent. Full-FP16-everything instead drops to ~24 dB through ODE
trajectory divergence
(not
degradation) — this mixed split keeps measurable fidelity while still putting the vocoder on the ANE.
**
189 MB** vs ~380 MB for FP32.
This repo is self-contained: the four
.mlpackage
graphs + the G2P-free tokenizer assets
(
tts.json
,
unicode_indexer.json
) +
voice_styles/
. The host runs the flow-matching ODE loop
(
vector_estimator
×
total_step
); the graphs contain no control flow.
Supertonic-3-CoreML-FP16 huggingface.co is an AI model on huggingface.co that provides Supertonic-3-CoreML-FP16's model effect (), which can be used instantly with this aufklarer Supertonic-3-CoreML-FP16 model. huggingface.co supports a free trial of the Supertonic-3-CoreML-FP16 model, and also provides paid use of the Supertonic-3-CoreML-FP16. Support call Supertonic-3-CoreML-FP16 model through api, including Node.js, Python, http.
Supertonic-3-CoreML-FP16 huggingface.co is an online trial and call api platform, which integrates Supertonic-3-CoreML-FP16's modeling effects, including api services, and provides a free online trial of Supertonic-3-CoreML-FP16, you can try Supertonic-3-CoreML-FP16 online for free by clicking the link below.
aufklarer Supertonic-3-CoreML-FP16 online free url in huggingface.co:
Supertonic-3-CoreML-FP16 is an open source model from GitHub that offers a free installation service, and any user can find Supertonic-3-CoreML-FP16 on GitHub to install. At the same time, huggingface.co provides the effect of Supertonic-3-CoreML-FP16 install, users can directly use Supertonic-3-CoreML-FP16 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
Supertonic-3-CoreML-FP16 install url in huggingface.co: