Ai

The Latxa model was an important milestone for artificial intelligence in the Basque language. Until then, we had no open-weight model that handled Basque decently. Since then, several others have been created specifically fine-tuned for Basque: Kimu (Orai) and the Latxa Qwen3 VL family of models recently released by HiTZ. At the same time, new top-tier local models have been emerging — Qwen, Gemma, etc. — which can be especially interesting for use in agents with external tools. These models are also improving their Basque language capabilities noticeably. But do they match the Latxa model? ...

Traditional Text-to-Speech technology gives us solid voices today, but offers no way to guide the interpretation of the audio. If we want to use these models to produce audiobooks, stories, game audio, etc., we soon realize the results are not ideal. We get cold, neutral voices. Robotic. Boring. Lately, however, models that allow us to specify paralinguistic cues (pauses, breathing, and emotional variations) are beginning to emerge. Thanks to this technology, by providing script-like instructions, we can create emotionally rich synthetic voices. ...

In recent years, text-to-speech (TTS) models have advanced enormously. Month after month, new releases improved pronunciation, prosody, and overall audio quality. Unfortunately, most of those models were designed for English, or at best compatible with only a few languages. The tools available for generating Basque voices used to be quite limited: paid proprietary systems (such as Elhuyar’s neural TTS, developed by Orai), or older robotic voices that had clearly fallen behind. ...

Over the last few days, I have been fine-tuning Nvidia Parakeet for Basque, and I have published the result here. The goal was simple: to have a lightweight Basque speech-to-text model that runs fast and is practical on modest hardware. Accuracy vs speed To be clear, this model is not as accurate as my best Basque Whisper model: xezpeleta/whisper-large-v3-eu However, it has a major advantage: it is very fast and can run on CPU-only setups. ...

Recently, the HiTZ center published the Latxa VL models. These models are based on Qwen3-VL models and possess vision capabilities. Currently, two sizes are available, both very small: models with 2B and 4B parameters. The tests presented below were performed with the 4B model. Models that know Basque Generally, models of such small size tend to only have the ability to answer decently in English. These Latxa VL models, however, have been trained to perform in Basque. ...