Tasks and inputs
neuphonic/neucodec-onnx-decoder-int8 is listed for Audio-to-audio, Neural Audio Codecs. It accepts audio codes and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
SyncDev model directory · Profile updated Oct 7, 2026
The NeuCodec INT8 ONNX decoder converts compatible NeuCodec codes into an audio waveform. It is an exported decoder, not a complete text-to-speech model.
The specification score summarizes recorded features and limits. It is not a benchmark result.
Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.
Accepts
Produces
The NeuCodec INT8 ONNX decoder converts compatible NeuCodec codes into an audio waveform. It is an exported decoder, not a complete text-to-speech model.
The underlying codec documents 16 kHz input, 24 kHz reconstructed output and a compact single-codebook representation. For an exported decoder, confirm that the supplied codes match the expected encoding and tensor layout.
Measure the complete path, including preprocessing and moving tensors or buffers between runtimes. Listen to quiet speech, names, sibilants and difficult recordings. A smaller representation or exported graph does not guarantee that every device will meet the same latency target.
The full model, ONNX decoder and INT8 decoder are related deployment artifacts. Keep them distinct when comparing: a decoder-only export cannot replace the encoder or the language model that generates speech codes. The checkpoints are under Apache 2.0 and have gated download conditions on Hugging Face.
Browse neural audio codecs and inspect the publisher's model card before selecting a runtime.
Compare the model on the work you need to complete.
neuphonic/neucodec-onnx-decoder-int8 is listed for Audio-to-audio, Neural Audio Codecs. It accepts audio codes and produces audio. Use the publisher’s input format and preprocessing for your chosen deployment.
Compare pronunciation, voice consistency and time to the first playable audio. Check the service’s billing unit for the actual script.
The weights are downloadable under Apache-2.0. Check runtime support and size the complete model and working context for your expected concurrency.
These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.
neuphonic
NeuCodec is Neuphonic's neural speech codec: it encodes audio into FSQ codes and reconstructs a waveform.
neuphonic
The NeuCodec ONNX decoder converts compatible NeuCodec codes into an audio waveform. It is an exported decoder, not a complete text-to-speech model.