Detail publikace

Phoneme Recognition using Temporal Patterns

MATĚJKA, P., SCHWARZ, P., ČERNOCKÝ, J., HEŘMANSKÝ, H.

Originální název

Phoneme Recognition using Temporal Patterns

Anglický název

Phoneme Recognition using Temporal Patterns

Jazyk

en

Originální abstrakt

We investigate and compare several techniques for automatic recognition of unconstrained context-independent phoneme strings from TIMIT and NTIMIT databases. Among the compared techniques, the technique based on TempoRAl Patterns (TRAP) achieves the best results in the clean speech, it achieves about 10% relative improovements against baseline system. Its advantage is also observed in the presence of mismatch between training and testing conditions. Issues such as the optimal length of temporal patterns in the TRAP technique and the effectiveness of mean and variance normalization of the patterns and the multi-band input the TRAP estimations, are also explored.

Anglický abstrakt

We investigate and compare several techniques for automatic recognition of unconstrained context-independent phoneme strings from TIMIT and NTIMIT databases. Among the compared techniques, the technique based on TempoRAl Patterns (TRAP) achieves the best results in the clean speech, it achieves about 10% relative improovements against baseline system. Its advantage is also observed in the presence of mismatch between training and testing conditions. Issues such as the optimal length of temporal patterns in the TRAP technique and the effectiveness of mean and variance normalization of the patterns and the multi-band input the TRAP estimations, are also explored.

Dokumenty

BibTex


@inproceedings{BUT7941,
  author="Pavel {Matějka} and Petr {Schwarz} and Jan {Černocký} and Hynek {Heřmanský}",
  title="Phoneme Recognition using Temporal Patterns",
  annote="We investigate and compare several techniques for automatic recognition of
unconstrained context-independent phoneme strings from TIMIT and NTIMIT
databases. Among the compared techniques, the technique based on TempoRAl
Patterns (TRAP) achieves the best results in the clean speech, it
achieves about 10% relative improovements against baseline system.
Its advantage is also observed
in the presence of mismatch between training and testing conditions.
Issues such as the optimal length of temporal patterns in the
TRAP technique and the effectiveness of mean and variance normalization of
the patterns and the multi-band input the TRAP estimations, are also explored.",
  booktitle="In Proceedings of the conference TSD'2003. International Conference on Text Speech and Dialogue, TSD 2003",
  chapter="7941",
  year="2003",
  month="september",
  pages="198",
  type="conference paper"
}