Model reference · open weights

Sidon-CoreML

Sidon-CoreML is an open-weight audio or speech model from aufklarer, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.

Audio aufklarer 1 variants 2k downloads/mo
Request this model on EU hardware All served models Not on the shared API today — deployed on request.

About

What Sidon-CoreML is

Sidon — Core ML (speech restoration / dereverberation) On-device speech restoration (denoise + dereverberation + bandwidth restoration) for Apple Silicon, exported to Core ML (Neural Engine + GPU). Turns a noisy/reverberant clip into studio-quality 48 kHz speech — ideal for cleaning a voice-cloning reference before TTS, since it preserves speaker identity. Two-stage pipeline: Variants Total 246 M params (predictor 193.6 M + DAC vocoder 52.4 M). Output sample rate 48 kHz. int8 keeps the vocoder at FP16 (audio quality); only the predictor is palettized. Files Quality (no-reference MOS, 10 s clip) DNSMOS P.835 (SIG/BAK/OVRL, higher = better) and UTMOS (naturalness, 1–5): Restoration lifts OVRL 2.90 → 3.29 (driven by BAK 3.40 → 4.11 — reverb removed). Quantization is near-lossless on DNSMOS and speaker similarity; UTMOS shows a small naturalness cost (fp16 −0.09, int8 −0.17). Numbers are a single clip — average over a set for a definitive figure. Front-end The graphs take inputfeatures [1, T, 160] from the w2v-BERT 2.0 SeamlessM4T feature extractor (16 kHz input). The sequence length is fixed (T = 499 ≈ 10 s) — chunk longer audio in the runtime. The front-end and chunking are handled by speech-swift. Usage Use via the speech-swift Apple SDK, e.g.: Compute placement. The runtime places the two stages separately: the predictor on the Neural Engine, the DAC vocoder on the GPU, where its very wide convolutions load and run fastest. Override both with --compute-units ane|gpu|cpu|all, or per stage via predictorComputeUnits: / vocoderComputeUnits: in the Swift API. Source Exported from Sidon (sarulab-speech), checkpoint sidon-v0.1; paper arXiv:2509.17052. Base SSL encoder: facebook/w2v-bert-2.0; vocoder: DAC (descript-audio-codec). All components are MIT-licensed. Links - speech-swift — Apple (Swift) SDK - soniqo.audio — website - blog

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

Makeraufklarer
TypeAudio & music
Variants1
Runs withcoreml
Based onsarulab-speech/sidon-v0.1
Released2026-06-14
Popularity2k downloads / month
LicenceOpen weights

How it works

How audio & music work

Audio or textinputAudio modelrecognise / synthesiseText or audiooutputSpeech-to-text turns audio into text; text-to-speech and music models turn text into audio.

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

VariantParamsPrecisionVRAMFits 16 GBWeights
Sidon-CoreMLBF16Weights ↗

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys sidon-coreml for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (sidon-coreml below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -F model="sidon-coreml" -F file=@audio.mp3

Details

Languages, data & research

Languages

multilingual

Tags

coreml sidon speech-restoration dereverberation speech-enhancement voice-cloning on-device audio-to-audio multilingual

Papers

Licence

Open weights

Open weights under mit — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗

Sources

Weights & code

Want Sidon-CoreML on EU-owned hardware?

Request this model on EU hardware See what’s served now

Explore

More audio & music

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms