Model reference · open weights
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.
About
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
| Maker | aufklarer |
|---|---|
| Type | Audio & music |
| Variants | 1 |
| Runs with | coreml |
| Based on | sarulab-speech/sidon-v0.1 |
| Released | 2026-06-14 |
| Popularity | 2k downloads / month |
| Licence | Open weights |
How it works
Variants
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.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| Sidon-CoreML | — | BF16 | — | — | Weights ↗ |
Using it via the API
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
Licence
Open weights under mit — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗