Recognition of Emotions in Mexican Spanish Speech: An Approach Based on Acoustic Modelling of Emotion-Specific Vowels

Table 8

Classification performance of the ASR system for speech and emotion recognition (individual speakers from the speech corpus) across five iterations.

Speaker	Emotion	Iteration 1	Iteration 2	Iteration 3	Iteration 4	Iteration 5
MS1	Anger	100.00	100.00	100.00	100.00	100.00
	Happiness	66.67	66.67	66.67	66.67	66.67
	Neutral	100.00	100.00	100.00	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	75.00	78.75	80.00	80.00	80.00
	Phoneme accuracy	80.12	82.40	83.46	84.67	83.46
MS2	Anger	71.43	71.43	83.33	83.33	100.00
	Happiness	100.00	83.33	75.00	100.00	75.00
	Neutral	100.00	100.00	100.00	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	78.26	86.34	88.20	87.58	92.55
	Phoneme accuracy	80.79	88.35	90.02	88.80	93.65
MS3	Anger	83.33	83.33	100.00	83.33	100.00
	Happiness	100.00	100.00	100.00	100.00	75.00
	Neutral	100.00	100.00	100.00	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	91.30	89.44	98.76	93.79	92.55
	Phoneme accuracy	90.77	91.38	98.49	94.86	93.65
FS1	Anger	100.00	83.33	100.00	100.00	100.00
	Happiness	100.00	100.00	100.00	100.00	100.00
	Neutral	100.00	100.00	66.67	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	76.25	74.38	68.13	80.63	85.63
	Phoneme accuracy	82.28	81.83	75.83	83.63	88.14
FS2	Anger	62.50	83.33	85.71	66.67	83.33
	Happiness	100.00	100.00	100.00	100.00	100.00
	Neutral	100.00	100.00	100.00	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	74.38	82.50	85.63	88.13	87.50
	Phoneme accuracy	77.13	87.04	87.80	89.33	88.11
FS3	Anger	85.71	62.50	71.43	83.33	83.33
	Happiness	100.00	100.00	100.00	100.00	100.00
	Neutral	100.00	100.00	100.00	100.00	100.00
	Sadness	100.00	100.00	100.00	100.00	100.00
	Word accuracy	88.75	74.38	92.50	79.38	88.75
	Phoneme accuracy	89.33	79.57	92.23	80.79	90.40