Skip to Main content Skip to Navigation
Conference papers

On semi-supervised LF-MMI training of acoustic models with limited data

Imran Sheikh 1 Emmanuel Vincent 1 Irina Illina 1 
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : This work investigates semi-supervised training of acoustic models (AM) with the lattice-free maximum mutual information (LF-MMI) objective in practically relevant scenarios with a limited amount of labeled in-domain data. An error detection driven semi-supervised AM training approach is proposed, in which an error detector controls the hypothesized transcriptions or lattices used as LF-MMI training targets on additional unlabeled data. Under this approach, our first method uses a single error-tagged hypothesis whereas our second method uses a modified supervision lattice. These methods are evaluated and compared with existing semi-supervised AM training methods in three different matched or mismatched, limited data setups. Word error recovery rates of 28 to 89% are reported.
Complete list of metadata

Cited literature [27 references]  Display  Hide  Download
Contributor : Emmanuel Vincent Connect in order to contact the contributor
Submitted on : Friday, July 31, 2020 - 3:19:23 PM
Last modification on : Saturday, July 23, 2022 - 3:53:00 AM


Files produced by the author(s)


  • HAL Id : hal-02907924, version 1


Imran Sheikh, Emmanuel Vincent, Irina Illina. On semi-supervised LF-MMI training of acoustic models with limited data. INTERSPEECH 2020, Oct 2020, Shanghai, China. ⟨hal-02907924⟩



Record views


Files downloads