A framework for integrating heterogeneous sporadic knowledge sources into automatic speech recognition

Stefan Ziegler 1 Guillaume Gravier 2
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
2 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Heterogeneous knowledge sources that model speech only at certain time frames are difficult to incorporate into speech recognition, given standard multimodal fusion techniques. In this work, we present a new framework for the integration of this sporadic knowledge into standard HMM-based ASR. In a first step, each knowledge source is mapped onto a logarithmic score by using a sigmoid transfer function. Theses scores are then combined with the standard acoustic models by weighted linear combination. Speech recognition experiments with broad phonetic knowledge sources on a broadcast news transcription task show improved recognition results, given knowledge that provides complementary information for the ASR system.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [17 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00906348
Contributor : Guillaume Gravier <>
Submitted on : Tuesday, November 19, 2013 - 3:41:48 PM
Last modification on : Friday, November 16, 2018 - 1:21:50 AM
Document(s) archivé(s) le : Thursday, February 20, 2014 - 9:30:43 AM

File

paper-07.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00906348, version 1

Citation

Stefan Ziegler, Guillaume Gravier. A framework for integrating heterogeneous sporadic knowledge sources into automatic speech recognition. Workshop on Speech, Language and Audio in Multimedia, 2013, France. pp.37-42. ⟨hal-00906348⟩

Share

Metrics

Record views

835

Files downloads

134