Interesting links, 02/08/2026

EvolvingLMMs-Lab/NEO — Native Vision-Language Models

tenstorrent/tt-metal — TT-NN operator library, and TT-Metalium low level kernel programming model.

Baidu just dropped an open-source multimodal AI that it claims beats GPT-5 and Gemini: ERNIE-4.5-VL-28B-A3B-Thinking

5 open-source remote desktop tools prove that nobody should use TeamViewer anymore

Human brain cells on a chip learned to play Doom in a week

WAXAL - A Large-Scale Multilingual African Language Speech Corpus, dataset

rustfs/rustfs — RustFS is an open-source, S3-compatible high-performance object storage system supporting migration and coexistence with other S3-compatible platforms such as MinIO and Ceph.

A neural network for modeling human concept formation, understanding and communication

40,000-year-old signs show humans were recording information long before writing

TW-Sound580K: A Regional Audio-Text Dataset with Verification-Guided Curation for Localized Audio-Language Modeling

Focus Then Listen: An Empirical Study of Plug-and-Play Audio Enhancer for Noise-Robust Large Audio Language Models

Multi-Loss Learning for Speech Emotion Recognition with Energy-Adaptive Mixup and Frame-Level Attention

davanstrien/ocr-bench-britannica-results-qwen35-viewer

Discrete Audio Tokens More Than a Survey!, taxonomy

Sleepwalking/SHIRO — Phoneme-to-speech alignment toolkit based on liblrhsmm

danielcopper/wezterm-session-manager — Lua script enhancement for WezTerm that provides functionality to save, load, and restore terminal sessions

MFA - Phone groups

Speech-Omni-Lite: Portable Speech Interfaces for Vision-Language Models

SceneGen: Single-Image 3D Scene Generation in One Feedforward Pass

MV-RAG: Retrieval Augmented Multiview Diffusion

CaviraOSS/OpenMemory — Local persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc.

DeepSeek OCR

dreamtheater123/Awesome-SpeechLM-Survey

WiT: Waypoint Diffusion Transformers via Trajectory Conflict Navigation

ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model

CohereLabs

DashengTokenizer: One layer is enough for unified audio understanding and generation, models

Voxtral TTS Demo

microsoft/harrier-oss-v1-27b

T5Gemma-TTS Technical Report

RushOnline/midi2hydrogen

AffectSpeech: A Large-Scale Emotional Speech Dataset with Fine-Grained Textual Descriptions for Speech Emotion Captioning and Synthesis

powertab/powertabeditor

What is the Heilmeier Catechism?

There was a 'Moved Permanently' error fetching URL: 'https://x.com/hooeem/status/2030720614752039185'

The Snack Sound Toolkit

A Speech Recognition Extension to Snack

chdh/klatt-syn

40,000-year-old signs show humans were recording information long before writing

Java to Kotlin Conversion Comes to Visual Studio Code

zpforlove/AG-REPA — Official code for AG-REPA: Causal Layer Selection for Representation Alignment in Audio Flow Matching (ICML 2026). Unified TTS+TTA Flow Matching with BiT-C / LASP / FoG-A diagnostics.

AustinZhang/AG-REPA — AG-REPA model.

Principles and Practice of Deep Representation Learning

Qwen3.5: Towards Native Multimodal Agents

Zaneham/Booth — Open-source CUDA, Triton and HIP compiler targeting multiple GPU and CPU architectures.

QwenLM/Qwen3-TTS

SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition

Exploring Fine-Tuning of Large Audio Language Models for Spoken Language Understanding under Limited Speech Data

Inverse-Hessian Regularization for Continual Learning in ASR

Libation — A free, open-source application for downloading and managing your Audible audiobooks

Mbucari/AAXClean — Decrypt Audible aax and aaxc files.

Qwen/Qwen3-ASR-1.7B

LESS: Large Language Model Enhanced Semi-Supervised Learning for Speech Foundational Models Using in-the-wild Data