GET IN TOUCH
GET IN TOUCH
Have something in mind? Let’s make it happen.
Have something in mind?
Let’s make it happen.
PRODUCTS
SOLUTIONS
COMPANY
Neuphonic © 2026
Train your own SpeechLMs
without needing to build your own codec
Train your own SpeechLMs
without needing to build your own codec
Access now!
Low bit-rate compression
Low bit-rate compression
A speech codec that compresses and reconstructs audio with near-inaudible reconstruction loss
A speech codec that compresses and reconstructs audio with near-inaudible reconstruction loss
Commercial use
Commercial use
Commercial use permitted
Commercial use permitted
Upsampling
Upsamples from 16kHz to 24kHz
Upsamples from 16kHz to 24kHz
Datasets
Released with large pre-encoded datasets.
Compressed Emilia-YODAS from 1.7TB to 41GB
Released with large pre-encoded datasets.
Compressed Emilia-YODAS
from 1.7TB to 41GB
Released with large pre-encoded datasets. Compressed Emilia-YODAS from 1.7TB to 41GB
NeuCodec is a lightweight neural codec that encodes audio at just 0.8 kbps. Perfect for researchers and builders who need something that just works for training high quality text-to-speech models.
NeuCodec is a lightweight neural codec that
encodes audio at just 0.8 kbps. Perfect for researchers
and builders who need something that just works
for training high quality text-to-speech models.
NeuCodec is a lightweight neural codec that
encodes audio at just 0.8 kbps. Perfect for researchers
and builders who need something that just works
for training high quality text-to-speech models.
KEY FEATURES
KEY FEATURES
NEUCODEC
NEUCODEC
NEUCODEC
An open-source neural codec framework for training speech LMs
An open-source neural codec
framework for training speech LMs
An open-source neural codec
framework for training speech LMs
Products
Solutions
Company
Products
Solutions
Company
ENTERPRISE
PRODUCTS
COMPANY
Neuphonic © 2026
ENTERPRISE
PRODUCTS
COMPANY
Neuphonic © 2026