rvq

NeuroRVQ: Multi-Scale Biosignal Tokenization for Generative Foundation Models

KAE : KAN-based AutoEncoder (AE, VAE, VQ-VAE, RVQ, etc.)

Unofficial PyTorch implementation of Higgs Audio V2 Tokenizer with HuBERT semantic features. Complete training pipeline for semantic-acoustic audio tokenization with 960x downsampling and 8-layer RVQ.

On the Limits of Discrete Representations for Neural Control. A systematic empirical study of tokenization, quantization, and inductive bias in BCI (aka documented failures)

📈 Model training, Pre-Training and Fine-Tuning workspace, scaling from a single RTX 3090 (24GB).

First-of-its-kind MIDI PCA RVQ VAE implementation and models

Residual vector quantization audio codec with a NumPy core: low-bitrate tokens for audio LMs, streaming encode/decode

NeuroRVQ biosignal tokenizer — pure Rust inference for EEG/ECG/EMG with Burn ML

Vectorial language for digital consciousness - RVQ-based emotion encoding (DeepSeek R1 validated)

Offline removal of neural-codec (RVQ) quantization artifacts from AI-generated music - comb suppression, frozen-noise unfreezing, optional Demucs neural stage

Build pure-Rust NeuroRVQ inference to tokenize EEG, ECG, and EMG signals with Burn 0.20 and zero Python dependencies

Vector-quantization compiler for LLM weights. Fits per-tensor RVQ codebooks, stores indices as bit-planes (~2 bits/weight), recovers quality with QAT, and exports to GGUF or vLLM.

End-to-end spoken dialogue toolkit: trainable RVQ neural audio codec, unified speech-text tokens, compact decoder-only LM with joint losses, KV-cache streaming generation, offline-tested adapter protocol for external speech LLMs, turn-level evaluation. NumPy core, CPU-only torch extra, synthetic data.

AI-powered ultra-low bitrate audio codec CLI built on SNAC — Compress audio at 0.98-2.6 kbps with near-original quality (392:1 ratio vs WAV)

Measured, verifiable removal of AI-music codec artifacts (RVQ residuals, sizzle, metallic ringing) - spectral gating + Demucs-guided restoration

Industrial-grade DSP, psychoacoustics & neural audio codec engine authored 100% natively in TokenVector (.tkv). 3-8 kbps RVQ, Chebyshev exciter, zero-GC execution

Open RVQ encoders for MiniMax Music 3 (v1–v4 + invert-and-project v5)

Pytorch implementation of a Moshi-inspired audio LM based on multiple backbones, inlcuding Qwen and a transformer decoder.

Directable multi-speaker speech synthesis: word-level tone/emotion/voice control, natural-language direction, RVQ audio codec (zero-dependency reference).

APU-Codec: Neural audio codec from source, optimized for AMD APU tri-processor inference (NPU encoder, GPU decoder)