WideCodec is a 44.1 kHz decoder finetune of neuphonic/neucodec.

Everything needed to run inference is in this repo — the neucodec/ package and infer_widecodec.py. No other source code required.

Or load it directly in Python:

Offline benchmark on a heavily multilingual 9,291-clip wideband set (50 clips × 188 clean ≥44.1 kHz datasets, 400+ language/dialect tags), faster-UTMOSv2 + spectral vs ground truth.

Bitrate = frame-rate × Σ log₂(codebook size). WideCodec's FSQ is levels=[4]×8, num_quantizers=1 → 8·log₂4 = 16 bits/frame × 50 frames/s = 800 bps = 0.8 kbps (vs DAC's 9 codebooks × 10 bits × 86 fps ≈ 8 kbps).

The 2026-08-16 release reaches UTMOSv2 2.919 on the 9,291-clip set — above ground truth (2.822) and level with NVIDIA NeMo-44k (2.903), at ~1/10 of NeMo's bitrate. It gained +0.131 UTMOSv2 over the previous release at unchanged content accuracy: CV22 macro CER is 7.70% for both (micro 7.09%), well inside the measured run-to-run spread of ±0.25 pt.

Same architecture, same losses, same 50 tokens/s and single codebook — only the training corpus changed, to clean DNSMOS-gated ≥44.1 kHz audio. The gain is not an artifact of the benchmark pool: on a held-out set of 1,330 clips from 138 sources absent from training, it scores 2.807 vs the previous release's 2.654 (+0.153, paired t=+14.2), and reference-based PESQ-WB also improves (+0.017, t=+7.5) — so naturalness and fidelity moved together rather than trading off. Full report: evaluation/REPORT_44k_class_codecs.md.

This model is trained in two stages. main always points at the latest stage.

Diverse, predominantly high-sample-rate human speech (no synthetic TTS). The scale-up group is SR-verified ≥ 44 kHz (probed before inclusion); the base corpora are long-form conversational, expressive and anechoic speech. Two groups:

To maximise acoustic diversity for the decoder (language-agnostic — the codebook is frozen), training is scaled with 336 additional datasets (162 at 48 kHz, 174 at 44.1 kHz), mirrored as <name>*audio.zip in malaysia-ai/Multilingual-TTS and SR-verified ≥ 44 kHz. Original sources:

See the canonical Load from Hugging Face section in the source repo. decoder_depth=20 is required — the weights are a depth-20 decoder, so loading with any other depth mismatches the architecture.

The discrete codes are identical to base neuphonic/neucodec — only the reconstruction sample rate and fidelity differ.