Parselmouth & Praat
The pipeline produces population_lingualab_audio
which is an implementation of Parselmouth Praat Scripts in Python by David R. Feinberg
This script will measure pitch, standard deviation of pitch, hnr, jitter, shimmer, and formants in python using parselmouth, a package that runs Praat in python.
Links to some relevant praat documentation
- Formant: Formant-manual
- Jitter: Jitter-manual
- Shimmer: Shimmer-manual
- PointProcess: PointProcess-manual
Variable definition
Characteristic/feature family | Definition |
---|---|
Fundamental Frequency (F0) | Variations in voice pitch, useful for analyzing intonation and emotional stress. |
Intensity (Volume) | Measure of loudness, can reflect emphasis or emotional changes. |
Voice Quality | Characteristics such as roughness, breathiness, clarity or nasality of the voice. |
Spectrogram | Visual representation of frequency evolution over time, useful for detailed analysis of sound characteristics. |
Formant Frequencies (F1, F2, etc.) | Important information on the characteristics of pronounced vowels. |
Frequency and Amplitude Modulation | Variations in frequency and amplitude within the same phoneme or word. |
Harmonic-to-Noise Ratio (HNR) | Ratio between harmonic and noisy components of the voice, useful for assessing vocal quality. |
Jitter (Frequency Variability) | Variability in the fundamental frequency of a sound, can indicate vocal stress or tension. |
Shimmer (Amplitude Variability) | Variability in the amplitude of sound waves, related to vocal stability and quality. |
Phonation | These characteristics focus on aspects of vocal production linked to vocal cord activity. They include measures such as pitch, jitter and shimmer, which are indicators of the frequency and amplitude stability of the voice. |
Articulation | These characteristics relate to the way words and sounds are formed by the speaker. They can include measures of phoneme duration, articulation speed and articulatory movement dynamics. |
Prosody | Prosody concerns the rhythm, intonation and accentuation of speech. Prosodic features include measures of intensity, syllable duration and stress patterns. |
Phonology | These characteristics analyze aspects of speech related to phonemic structure, such as changes in vowel or consonant quality. |
Variable names
F0_std
duration
f0_mean
formants_1_mean
formants_1_median
formants_2_mean
formants_2_median
formants_3_mean
formants_3_median
formants_4_mean
formants_4_median
hnr
jitter_ddp
jitter_local
jitter_local_absolute
jitter_ppq5
jitter_rap
shimmer_apq11
shimmer_apq3
shimmer_apq5
shimmer_dda
shimmer_local
shimmer_local_dB