Models listed
3
Published in this category
Worth weighing up
What goes in and out, licence terms, and whether you can host it yourself
How we treat data
Published specs only — we don't estimate numbers a vendor hasn't stated
Neural audio codecs convert audio into a compact learned representation and reconstruct a waveform from those codes. A codec can be one stage of a speech-language or text-to-speech system; a decoder alone is not a complete speech generator.
Check input and output sample rates, token rate, codebook format and runtime compatibility. Listen to difficult clips as well as clean speech and measure end-to-end latency on the intended device.
Explore NeuCodec, its ONNX decoder and the INT8 decoder. Treat the exported decoder variants as deployment options for a related codec rather than unrelated general audio models.
Scan the table below, then open any model for pricing, limits and the full write-up.
| neuphonic/neucodec neuphonic | 10.0 | — | — | — | — |
| neuphonic/neucodec-onnx-decoder neuphonic | 10.0 | — | — | — | — |
| neuphonic/neucodec-onnx-decoder-int8 neuphonic | 10.0 | — | — | — | — |
1–3 of 3 models · select a column to sort
About this order: Recommended shows our editorial shortlist first, followed by recent releases. The spec score is a 0–100 figure SyncDev calculates from each vendor's published specifications — context window, reasoning support, input modalities, tool calling, maximum output and how recently the model shipped. It measures breadth of capability, not benchmark performance, so a higher-scoring model is not automatically the better choice for your task.