Sentence examples for spontaneous database from inspiring English sources

For the spontaneous database, the values decreased to 88 (SFS) and 90 (FMS) from the initial 473.

Among a number of databases, we have used well-known EMO-DB [15] and VAM [16] as, respectively, acted and spontaneous database.

As a result, these spontaneous databases present challenges in analysis, because there is no defined population from which to base the denominator when estimating reporting rates.

In addition, the spontaneous speech database chosen for the experimentation, and in particular its realistic and challenging acoustic conditions, made the evaluation and the database attractive enough for future research.

Significantly higher results can be seen for spontaneous speech database.

For the purpose of the examination, a Polish spontaneous emotion database was created.

The evaluation is conducted on a Spanish spontaneous speech database, which comprises a set of talks from workshops and amounts to about 7 h of speech.

Evaluation is conducted on a Spanish spontaneous speech database containing a set of talks from workshops, which amount to about 7 h of speech.

Evaluation is conducted on a Spanish spontaneous speech database containing a set of talks from MAVIR workshopsb, which amount at about 7 h of speech in total.

An improvement of recognition can be also observed after applying FMS, with the exception of f 1−f 3 and LPC in the case of acted speech and BFCC and PLP in the case of the spontaneous speech database.

Figure 8 shows spectrograms, fundamental frequency (F0), F1F3syn and breathiness power contours for representative speech utterances, found in our spontaneous speech database, where br/wh segments (indicated by arrows) appear in the expression of several paralinguistic information items.