Sentence examples for commentators speech from inspiring English sources

In section 3, analysis of speech production features and their correlation with excitement in commentators' speech in the context of sports videos is discussed.

In this section, we first evaluate the speech production-based classifier in terms of the excitability measure for sports commentators speech.

Inspired by studies on emotion assessment [18 20], in this study, we analyze the effects of excitement on the parameters of a linear model of speech production derived from commentators' speech.

The multi-modal events/features used for excitability measure include: (1) slow motion replay, (2) camera motion activity, (3) scene cut density, (4) commentators' speech in high and (5) low excitement levels, and (6) audio energy.

The multi-modal events/features used in the proposed highlights generation system are the following: slow motion replay, camera motion activity, scene cut density, and excitement in commentators' speech.

Experiments are performed on baseball videos based on signal processing advancements for excitement assessment in the commentators' speech, audio energy, slow motion replay, scene cut density, and motion activity as features.

Excitement level detected in commentator speech is another good indication of excitability (ρ=0.5146).

In such a case, the statistics of commentator's speech features may be estimated on-the-fly.

Figure 15 Impact of on-the-fly estimation of commentator's speech statistics on excitement level classification error.

The feature normalization utilized in the previous paragraph assumes that the whole game is available to estimate the global statistics of the commentator's speech (see section 3).

Based on the analysis of the sports commentator's speech, production parameters most correlated with the perceptual excitability were selected to form an acoustic feature vector for excitement assessment.