The Multilingual Hispanic Speech in California Corpus

Students being trained to collect speech data
4.7 million words
The MuHSiC (Multilingual Hispanic Speech in California) project is a corpus of bilingual Spanish-English speech collected through sociolinguistic interviews and naturalistic conversations across California.
Undergraduate students collected speech from participants with diverse social and regional backgrounds. Graduate students and the three faculty principal investigators trained the undergraduate researchers.
The corpus contains 1,200 recordings of approximately 35 minutes each: 600 recordings in Spanish and 600 in English.
The Multilingual Hispanic Speech in California Corpus has been made possible in part by a Multicampus Research Programs and Initiatives Grant (MRPI Grant M23PL5866) awarded to Mark Amengual, Ji Young Kim, and Justin Davidson. Any views, findings, conclusions, or recommendations expressed on this website or through this project do not necessarily represent those of the University of California.

Project Team

Contact

For inquiries about corpus processing, downloads, web development, or dataset access, contact julianvargo at berkeley.edu