Latest AI and tech news
Drama 3 architecture. Shapes and parameter counts from config.json and the safetensors headers; flow from arXiv 2610.18010....
For the past two weeks, an AI agent ran autonomous architecture research on BlaGPT, my GPT-2-scale playground for testing LLM techniques. It ran about 100 short proxy experiments and 22 full training runs, implemented techniques from papers, kept its...
Yann LeCun gave a talk at ETH Zürich last week, “World Models: Enabling the next AI revolution”. I went through the whole thing and checked his claims against the actual papers. Here is where he is right, where he overreaches....
Efficient large language models have driven various architectural innovations—from mixture-of-experts to quantization techniques. Attention mechanisms remain the core computational bottleneck in transformers, and optimization typically degrades outpu...
MiMo-Audio is Xiaomi’s 7B parameter model that processes speech and text through a unified architecture. Trained on 100+ million hours of audio data—10x larger than existing open-source models— that results in emergent capabilities like voice convers...
The landscape of text-to-speech synthesis has been dominated by a persistent trade-off: generate short, high-quality utterances, or attempt longer sequences with degraded coherence. While recent models have pushed the boundaries of single-speaker syn...
The race for more capable large language models has largely been about scaling—bigger models, more parameters, more compute. NVIDIA’s Nemotron Nano 2 challenges this paradigm with a hybrid architecture that achieves competitive performance through ar...
Text-to-Speech (TTS) systems have traditionally struggled with the trade-off between quality and latency. Most high-quality systems require processing the entire text before generating audio, while streaming approaches often sacrifice naturalness. Ky...
Speech synthesis has evolved rapidly, but most systems still struggle with natural expressiveness and multi-modal understanding. Higgs Audio V2 addresses these limitations through a unified architecture that treats audio as a language, enabling emerg...
Recently, I’ve been testing different architectural modifications in my BlaGPT benchmark to see how they affect performance. I wanted to share what I’ve learned about three interesting techniques: Softpick, Canon Layers, and Parallel Transformer bloc...
When working with language models for audio tasks, switching between different audio representations can be challenging.
Let’s say you have a text-to-speech model using a particular audio codec, and a new codec comes along with better quality.
The ob...
Note: Read and think a lot, be a researcher. Build the expert intuition. I’m sure you get stuck at some point without it....
Having dedicated the last half-dozen years to an open-source AI, here are my insights on the pros and cons of operating as an open-source organization....
Here I wrote a short note on my experiment with training XTTS for Voice Conversion....
We recently released XTTSv2 with 🐸TTS v0.20, and here I go over the relevant details of the model....
The XTTS v1.1 version now includes Japanese, along with some model enhancements and a fresh vocoder. As a result, this iteration of XTTS can now speak in 14 languages with improved quality,...
In this article, I will explore various alternatives to transformers, considering their architectural improvements, computational efficiency, and performance results across different benchmarks. I intend to continually update this post with new model...
XTTS is a versatile Text-to-speech model that offers natural-sounding voices in 13 different languages. One of its unique features is the ability to clone voices across languages using just a 3-second audio sample....
We recently open-sourced XTTS, the best open-source text-to-speech model we’ve released so far....
👉 Subscribe to my Substack to get the latest news and articles....
Abstracting ML code sacrifices expressiveness, increases coupling, and aggravates maintenance. These might be ok for regular software. But things are different for ML. I am sure you know how it feels to waste hours trying to match the API when you wa...
The recent surge of new end-to-end deep learning models has enabled new and exciting Text-to-Speech (TTS) use-cases with impressive natural-sounding results. However, most of these models are trained on massive datasets (20-40 hours) recorded with a ...
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....
This post was originally published on erogol.com (hosted on DigitalOcean) and recovered from the Wayback Machine after the original server was lost. Some formatting or images may differ from the original....