This work aimed to estimate the welfare of pigs using their vocalization records during rearing in a commercial swine farm. A total of 150 barrow pigs were randomly distributed into five pens. Pigs were exposed to different stressful situations: Thirst (no access to water), hunger (no access to the food), thermal stress (Temperature and Humidity Index - THI >74). For the control treatment, the animals were kept in a situation of comfort, with full access to food and water, and environment THI <70. Acoustic signals were recorded every 30 minutes during a continuous period of three hours, totaling six samples for each stress exposure. The signals were digitized at a frequency of up to 44,100 Hz, for a period of 3 minutes. Later the audios were analyzed using the Praat (r) 5.1.19 software. The attributes generated from this software were the signal energy (Pa² * s), the maximum amplitude and the minimum amplitude (Pa), the frequency of pitch (Hz), the sound intensity (dB) and four levels of formants (F1, F2, F3 and F4), also called harmonics (Hz). It was found that depending on the acoustic attributes, the stressor stimuli and its duration indicate energy and signal strength, frequency of pitch and formants 2 and 4 showed differentiation. The sound attributes of the pig vocalization varied in different ways depending on the type and duration of the stressful exposure, functioning as an efficient tool to quantify the degree of animal degree.
ambience; stress; pig farming; precision livestock production
