Echo delay magnitude in telecommunication maps to distinct perceptual degradations.

Evidence

A common unwanted feature of two-way telecommunication is the presence of loud echoes (Appel and Beerends, 2002). In natural conditions, a speaker perceives his or her own voice from the mouth-to-ear path and from the acoustic response of the environment, providing direct feedback that is used subconsciously to control the speech production process. Interference in the feedback can influence the comfort of speaking and also the manner of speaking. A wellknown effect is the raising of one’s voice in the presence of loud background noise, called the Lombard effect (Lane and Tranel, 1971). In contrast, we lower the volume of our voice when we are played back our voice loudly. Delaying the echo increases the perception of discomfort. For small delays (<10 ms) and high levels, the echo interferes with the sound coming directly from the mouth of the speaker, leading to the perception of colouration due to comb filtering. Medium delays (10–30 ms) lead to the perception of hollowness in one’s voice, and for larger delays (> 30 ms) we perceive a clear, distinct echo. When the delay is large (> 200 ms) and the level of the echo is high, subjects experience difficulty in producing words (Appel and Beerends, 2002). The difficulties may be simply because, when producing a syllable, one’s own voice present in the ear canals due to the echo has the timbre of the previous syllable, which causes confusion in our neural speech production system. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2459

Categorization

Category: Causal Relationships

  • DEFINE: Introduces the Lombard effect and comb-filtering colouration as technical terms.
  • XREF: Echo-feedback disruption of speech control connects to prior chapters on how self-vocalization feedback shapes production.