Dear Diarization: Using MFA 2.1 to assign speaker labels to utterances
Evaluating speaker diarization in MFA for English and Korean
Playing around with PolyglotDB
Following some updates to PolyglotDB, let’s play around with some recordings and extract some acoustic measures
How much data do you need for a good MFA alignment?
Forced alignment is sometimes more art than science, but let’s try to establish some rules of thumb for how much data you might need when aligning a corpus
Speaker dictionaries and multilingual IPA
Let’s take a closer look at some new features in MFA 2.0, multilingual IPA and speaker-specific dictionaries and how they can be used to align multi-dialectal and multilingual corpora
Update on Montreal Forced Aligner performance
Take a look at how MFA 2.0 stacks up against the first version and a couple of other aligners in English
Bootstrapping an IPA dictionary for English using Montreal Forced Aligner 2.0
Want to bootstrap and clean up a new dictionary? This tutorial covers my experience creating a new IPA dictionary for English using Montreal Forced Aligner’s 2.0 functionality
How to create synthetic speech continua using STRAIGHT
STRAIGHT is a program that analyzes speech, decomposes it into source and filter characteristics, and then recombines it after some manipulation or interpolation. This tutorial is more of a practical guide for using STRAIGHT to create various synthetic speech