Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
These are the NeuCodec decoder weights, extracted to safetensors for use with a pure-Rust CPU decoder (ndarray + rustfft) — no native ONNX runtime, no PyTorch at inference.
Model source
Source description
These are the NeuCodec decoder weights, extracted to safetensors for use with a pure-Rust CPU decoder (ndarray + rustfft) — no native ONNX runtime, no PyTorch at inference.
The decoder converts the discrete speech codes produced by the NeuTTS backbone back into a 24 kHz audio waveform, completing the zero-shot voice cloning pipeline.
This artifact exists to enable voice cloning in , a local-first desktop AI application, under strict architectural constraints: no native ONNX runtime (no / ), no GPL dependencies. The companion encoder is .
Sources
1 sourceVerified Sep 19
Model artifacts
1 artifactSource excerpts
2 excerpts
onnxruntimeortsafetensors (extracted from the original pytorch_model.bin)Upstream distributes the decoder only as ONNX (neuphonic/neucodec-onnx-decoder); the pure-Rust decoder loads safetensors. These weights were extracted from the original PyTorch model:
pytorch_model.bin from neuphonic/neucodecsafetensors (script: convert_weights.py, or the dependency-free convert_weights_nopytorch.py, in the Ragtag repo).No weights are modified — this is a format conversion only.
The pure-Rust decoder was verified against the reference implementations:
NeuCodec is a generative (vocoder) decoder, so two valid implementations produce intelligible but not phase-identical waveforms; intelligibility, not waveform correlation, is the meaningful check.
This is not a standalone TTS system. It is the decoder stage of Ragtag's clone pipeline, used together with:
The decoder runs on CPU.
Licensed under Apache License 2.0, derived from the original NeuCodec which is also Apache 2.0 licensed. Please retain attribution to the original authors:
NeuCodec by Neuphonic
https://huggingface.co/neuphonic/neucodec
Licensed under Apache 2.0
The safetensors extraction is contributed by Ragtag / Captivated Ltd, also under Apache 2.0.
@misc{neucodec2024,
author = {Neuphonic},
title = {NeuCodec},
year = {2024},
publisher = {Hugging Face},
url = {https://huggingface.co/neuphonic/neucodec}
}
@misc{neucodec-decoder-safetensors,
author = {Mallett, Leon},
title = {NeuCodec Decoder (safetensors weights)},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/ragtag-ai/neucodec-decoder}
}
--- license: apache-2.0 language: - en tags: - audio - speech - text-to-speech - voice-cloning - neucodec library_name: safetensors base_model: neuphonic/neucodec inference: false --- # NeuCodec Decoder (safetensors weights) These are the [NeuCodec](https://huggingface.co/neuphonic/neucodec) decoder weights, extracted to `safetensors` for use with a pure-Rust CPU decoder (ndarray + rustfft) — no native ONNX runtime, no PyTorch at inference. The decoder converts the discrete speech codes produced by the NeuTTS backbone back into a 24 kHz audio waveform, completing the zero-shot voice cloning pipeline. This artifact exists to enable voice cloning in [Ragtag](https://ragtag-ai.app), a local-first desktop AI application, under strict architectural constraints: no native ONNX runtime (no `onnxruntime` / `ort`), no GPL dependencies. The companion encoder is [ragtag-ai/neucodec-encoder-rten](https://huggingface.co/ragtag-ai/neucodec-encoder-rten). ## Model details - **Source model:** [neuphonic/neucodec](https://huggingface.co/neuphonic/neucodec) (decoder portion only) - **Format:** `safetensors` (extracted from the original `pytorch_model.bin`) - **Runtime:** a pure-Rust CPU decoder (ndarray + rustfft) — the implementation in [neutts-rs](https://github.com/eugenehp/neutts-rs), vendored into Ragtag - **Precision:** fp32 - **File size:** approximately 880 MB - **Input:** NeuCodec speech codes (50 per second of audio) - **Output:** 24 kHz mono audio waveform ## Provenance and extraction process Upstream distributes the decoder only as ONNX ([neuphonic/neucodec-onnx-decoder](https://huggingface.co/neuphonic/neucodec-onnx-decoder)); the pure-Rust decoder loads `safetensors`. These weights were extracted from the original PyTorch model: 1. **PyTorch source:** `pytorch_model.bin` from [neuphonic/neucodec](https://huggingface.co/neuphonic/neucodec) 2. **Extraction:** the decoder sub-graph (VocosBackbone: FSQ projection, conv embed, ResNet + RoPE transformer blocks, ISTFT head) is filtered out and written to `safetensors` (script: `convert_weights.py`, or the dependency-free `convert_weights_nopytorch.py`, in the Ragtag repo). No weights are modified — this is a format conversion only. ## Quality verification The pure-Rust decoder was verified against the reference implementations: - **Decoder parity:** 0.99 correlation with the upstream Rust decoder...
Source context: 0 downloads · 0 likes · Pipeline text-to-speech · Library safetensors · Repo ragtag-ai/neucodec-decoder