STFT-VAE is a continuous VAE neural audio codec that uses STFT, transformer encoder-decoder, and ISTFT to achieve a very low latent frame rate (3.125 Hz) while preserving phase and outperforming SNAC on objective metrics.
43
No data
2
0
MIT
2026-07-12
It achieves 4–15× lower frame rate than comparable codecs (Mimi, SNAC, MioCodec) yet outscores SNAC on all tested metrics (SI-SDR, PESQ, STOI), and it preserves phase directly through STFT/ISTFT without a learned vocoder.
Inference for the STFT-VAE continuous audio codec (24kHz, 3.125Hz latent)
Similar projects matched by category, topics, and programming language.
Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.
mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.