Speech Production Dataset for Canavesano Piedmontese

This study presents the first comprehensive acoustic analysis of the vowel system in Canavesano Piedmontese, an endangered variety of Piedmontese spoken in the Canavese subregion of northwest Italy. Nineteen native speakers aged 57-83 participated in sociolinguistic interviews, generating a dataset of 2426 tokens analyzed for formant frequencies (F1 and F2). Results confirm the presence of nine vowels /a,ɛ,e,ø,ə,y,o,u,i/, including the mid-central vowel /ə/, which exhibits distinct acoustic properties compared to surrounding vowels. Despite individual variability among speakers, the community-level analysis reveals a stable vowel system similar to nearby Turinese Piedmontese. These findings contribute to the documentation and preservation of Canavesano Piedmontese, highlighting the importance of acoustic phonetic studies in understanding and conserving endangered languages.