Speech-to-text

Basque from speech to text: improving transcription models

Speech-to-text (STT) technology converts spoken language into written text using natural language processing. These systems are becoming increasingly important in digital interfaces, accessibility solutions, and various communication platforms. Since 2022, I’ve been fine-tuning the original Whisper speech recognition model for the Basque language, using the Mozilla Common Voice dataset. Compared to the original models, I’ve seen significant improvements in performance. As the Mozilla Common Voice initiative has grown, the model’s accuracy has continued to improve. ...

February 27, 2024
GitHub Copilot Agent Mode

GitHub Copilot Agent Mode: An Intelligent Development Partner

GitHub Copilot has introduced a significant advancement: Agent Mode. With this new feature, using the VS Code editor’s text interface, it can autonomously analyze, modify, and execute files in your project. What is Agent Mode or SWE agent? Agent Mode expands GitHub Copilot’s capabilities, transforming it from a simple code completer to an autonomous development assistant. The artificial intelligence can analyze, understand, and modify project files, as well as execute terminal commands. ...

February 12, 2024
Tulu-3 workflow

Local RAG in Basque using Tülu 3

In recent months, Local LLMs have significantly improved, and some of them perform surprisingly well with Basque language. Among them, I want to highlight Tülu 3 70B, which shows good results in Basque when using the quantized version (q4_K_M). Until the Latxa instruction model becomes available, this is probably the best option for having conversations or generating text in Basque. ...

February 9, 2024

New blog

Hi! This is my new blog, where I will write about new things I learn. I know, “blog” looks an outdated term… maybe should I try with Tik Tok? 😂 ...

November 5, 2021