Update about.md

about.md

@@ -224,11 +224,9 @@ This version is tailored for the Catalan language, as it was trained only on Cat
 
 ## Adaptation to Catalan
 
-The original Matcha-TTS model excels in English, but to bring its capabilities to Catalan, a multi-step process was undertaken. Firstly, we fine-tuned the model from English to Catalan central (Matxa-base), which laid the groundwork for understanding the language's nuances. This first fine-tuning from English was done using two datasets:
-
- * [Our version of the openslr-slr69 dataset.](https://huggingface.co/datasets/projecte-aina/openslr-slr69-ca-trimmed-denoised) 
-  
- * [Our version of the Festcat dataset.](https://huggingface.co/datasets/projecte-aina/festcat_trimmed_denoised)
+The original Matcha-TTS model excels in English, but to adapt it to Catalan, we have carried out a multi-stage process. 
+First, we fine-tuned the English to Central Catalan model by creating a Matxa-base, using a 100h subset of the CommonVoice v.16 Catalan database. 
+The selection of this small set of samples has been performed automatically by using the UTMOS system, a predictor of values of the metric Mean Opinion Score (MOS) a score that is usually set by humans according to their subjective perception of the quality of the speech.
 
 Then we further fine-tuned the single accent Catalan Matxa-based model with the soon to be published LaFrescat dataset that has 8.5 hours of recordings for four dialectal variants: