MBROLA
Jump to navigation
Jump to search
The MBROLA project, started by the TCTS Lab of the Faculté Polytechnique de Mons (Belgium), provides speech databases in many languages.
Overview
The core application of the project is MBROLA, a free (for non-commercial/non-military use) synthesizer that concatenates diphones, using a set of phonemes (sets of speech sounds specific to a language) for input, along with prosodic data, rendering speech samples at sixteen bits. It is technically not a text-to-speech synthesizer, because it doesn't use raw text as input.
For details on how the project is organized, shared and developed, please visit the MBROLA home site
See the University of Edinburgh help page to use Festival with MBROLA.