222
D. Cotoros et al.
construction site noises or white noise is less annoying and leads to a small number of
mistakes while reading on pop music is rather a challenge, involving many mistakes
and higher intensity.
As far as frequency is concerned, human voice is of course not a pure sound; the
sound spectrum shows a larger range of values. The fundamental frequency for men’s
voice is usually between 85 and 180 Hz, while for women, it gets to 165–255 Hz, but
many harmonic series will be present and the impression of hearing the fundamental
tone. In fact, the usable voice frequency band ranges between 300 and 3400 Hz.
The investigated subjects started from a fundamental frequency of 172 Hz—the
male subjects and 210–258 Hz the female subjects. A 20% growth in frequency was
recorded when the subjects were asked to read with construction site and white noise.
The frequency increased up to 50% during reading with pop music.
4 Conclusions
There are many situations in real life when people need to speak to an audience
in a noisy environment due to faulty technical devices, large or bored audience,
construction sites or maintenance work nearby, traffic, music, etc. Human voice will
involuntarily change its characteristics when subjected to a noisy environment, and
this paper analyzed the characteristics during four different types of environments.
The results showed that all the subjects regardless if they had any knowledge about
Lombard’s effect displayed dramatic growth in both intensity and frequency of speech
when the perturbation involved pop music with human voice. Also the number of
reading mistakes was considerably higher.
As for the other types of sounds, construction site and white noise, a slightly higher
intensity and frequency were obtained and also just a few reading mistakes as the
subjects declared that impersonal noises are easier to be filtered by their minds than
noises involving human voices. We intend to further analyze the subject involving
discussion between two or more people in order to determine the influence of human
voice as a disturbing speech factor.
References
1. H. Brumm, S.A. Zollinger, The evolution of the Lombard effect: 100 years of psychoacoustic
research. Behaviour 148, 1173–1198 (2011)
2. D. Vlaj, Z. Kacic, The Influence of Lombard Effect on Speech Recognition (Speech Technologies,
Maribor, Slovenia, 2011), pp. 151–168
3. V.S. Varadarajan, J.H.L. Hansen, Analysis of Lombard effect under different types and levels of
noise with application to In-set speaker ID systems, in INTERSPEECH 2006-ICSLP, Pittsburgh,
Pennsylvania (2006), pp. 937–940
4. P. Bottalico, I.I. Passione, S. Graetzer, E. Hunter, Evaluation of the starting point of the Lombard
effect. Acta Acust. Unit. Acust. 103(1), 169–172(4) (2017)
D. Cotoros et al.
construction site noises or white noise is less annoying and leads to a small number of
mistakes while reading on pop music is rather a challenge, involving many mistakes
and higher intensity.
As far as frequency is concerned, human voice is of course not a pure sound; the
sound spectrum shows a larger range of values. The fundamental frequency for men’s
voice is usually between 85 and 180 Hz, while for women, it gets to 165–255 Hz, but
many harmonic series will be present and the impression of hearing the fundamental
tone. In fact, the usable voice frequency band ranges between 300 and 3400 Hz.
The investigated subjects started from a fundamental frequency of 172 Hz—the
male subjects and 210–258 Hz the female subjects. A 20% growth in frequency was
recorded when the subjects were asked to read with construction site and white noise.
The frequency increased up to 50% during reading with pop music.
4 Conclusions
There are many situations in real life when people need to speak to an audience
in a noisy environment due to faulty technical devices, large or bored audience,
construction sites or maintenance work nearby, traffic, music, etc. Human voice will
involuntarily change its characteristics when subjected to a noisy environment, and
this paper analyzed the characteristics during four different types of environments.
The results showed that all the subjects regardless if they had any knowledge about
Lombard’s effect displayed dramatic growth in both intensity and frequency of speech
when the perturbation involved pop music with human voice. Also the number of
reading mistakes was considerably higher.
As for the other types of sounds, construction site and white noise, a slightly higher
intensity and frequency were obtained and also just a few reading mistakes as the
subjects declared that impersonal noises are easier to be filtered by their minds than
noises involving human voices. We intend to further analyze the subject involving
discussion between two or more people in order to determine the influence of human
voice as a disturbing speech factor.
References
1. H. Brumm, S.A. Zollinger, The evolution of the Lombard effect: 100 years of psychoacoustic
research. Behaviour 148, 1173–1198 (2011)
2. D. Vlaj, Z. Kacic, The Influence of Lombard Effect on Speech Recognition (Speech Technologies,
Maribor, Slovenia, 2011), pp. 151–168
3. V.S. Varadarajan, J.H.L. Hansen, Analysis of Lombard effect under different types and levels of
noise with application to In-set speaker ID systems, in INTERSPEECH 2006-ICSLP, Pittsburgh,
Pennsylvania (2006), pp. 937–940
4. P. Bottalico, I.I. Passione, S. Graetzer, E. Hunter, Evaluation of the starting point of the Lombard
effect. Acta Acust. Unit. Acust. 103(1), 169–172(4) (2017)