Fast bootstrapping of LVCSR systems with multilingual phoneme sets

By Tanja Schultz, A. Waibel. 1997

In this paper we described an e cient method to bootstrap continuously spoken, large vocabulary speech recognition systems by multilingual phoneme sets. To evaluate this techniques we collected the multilingual database GlobalPhone which currently consists of 9 di erent languages. A multilingual recognizer (MULTI) based on the four languages German, English, Japanese and Spanish was developed to serve as a source system. Likewise this system is very useful for language identi cation and achieves 100% language identi cation rate. Based on the MULTI system we evaluated our bootstrap technique on such completely di erent languages as Chinese, Croatian, and Turkish.

Subfields: Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Speech & Audio

Citations in the corpus

Language-independent and language-adaptive acoustic modeling for speech recognition (2001, methodology): “Recently, we demonstrated the usefulness of a global unit set and multilingual acoustic models as seed models (Schultz and Waibel, 1997).”
Data augmentation for low resource languages (2014, background): “One group relies on the use of a universal phone set to accomplish mapping of one language to another [28, 3].”
Multilingual training of deep neural networks (2013, methodology): “Previous approaches in this direction have included: constructing a universal phone set [1, 2, 3]; modeling a set of universal speech attributes, such as voicing, nasality and frication [4]; and mapping between phones of different languages using some automated method that relies on some predeﬁned…”
Universal Phone Recognition with a Multilingual Allophone System (2020, background): “The first group, shared phoneme models, creates a shared phoneme inventory of all phonemes from all training languages [3, 4, 5, 6, 7, 10].”