Projects per year
In the recent years, different types of invariant features have been pro-posed which promise to improve the robustness of speech recognition systems inmismatching training-test conditions with respect to the mean vocal tract lengths.Many state-of-the-art systems make use of system combination. By consideringspeech recognition systems with different front ends, this paper investigates whetherthe system combination of standard-feature and invariant-feature based systemswith ROVER yields improvements in accuracy. Results show that the combina-tion of the considered systems leads to clear accuracy improvements. An erroranalysis also shows that the considered invariant features carry different types ofinformation than the standard ones. The performance achieved with our systemcombination is in the range of what the best systems achieve in literature, althoughour approach does not yet include discriminative training or heteroscedastic featuretransformation.
|Number of pages
|Published - 01.09.2011
|Elektronische Sprachsignalverarbeitung 2011 : Tagungsband der 22. Konferenz - Aachen, Germany
Duration: 28.09.2011 → 30.09.2011
|Elektronische Sprachsignalverarbeitung 2011 : Tagungsband der 22. Konferenz
|28.09.11 → 30.09.11
FingerprintDive into the research topics of 'Robust Continuous Speech Recognition through Combination of Invariant-Feature Based Systems'. Together they form a unique fingerprint.
- 1 Finished
01.01.07 → 31.12.11
Project: DFG Projects › DFG Individual Projects