Degraded and computer-generated speech processing in a bonobo

Abstract

The human auditory system is capable of processing human speech even in situations when it has been heavily degraded, such as during noise-vocoding, when frequency domain-based cues to phonetic content are strongly reduced. This has contributed to arguments that speech processing is highly specialized and likely a de novo evolved trait in humans. Previous comparative research has demonstrated that a language competent chimpanzee was also capable of recognizing degraded speech, and therefore that the mechanisms underlying speech processing may not be uniquely human. However, to form a robust reconstruction of the evolutionary origins of speech processing, additional data from other closely related ape species is needed. Specifically, such data can help disentangle whether these capabilities evolved independently in humans and chimpanzees, or if they were inherited from our last common ancestor. Here we provide evidence of processing of highly varied (degraded and computer-generated) speech in a language competent bonobo, Kanzi. We took advantage of Kanzi’s existing proficiency with touchscreens and his ability to report his understanding of human speech through interacting with arbitrary symbols called lexigrams. Specifically, we asked Kanzi to recognise both human (natural) and computer-generated forms of 40 highly familiar words that had been degraded (noise-vocoded and sinusoidal forms) using a match-to-sample paradigm. Results suggest that—apart from noise-vocoded computer-generated speech—Kanzi recognised both natural and computer-generated voices that had been degraded, at rates significantly above chance. Kanzi performed better with all forms of natural voice speech compared to computer-generated speech. This work provides additional support for the hypothesis that the processing apparatus necessary to deal with highly variable speech, including for the first time in nonhuman animals, computer-generated speech, may be at least as old as the last common ancestor we share with bonobos and chimpanzees.

LinkedOXTRVariants Are Associated with Social Behavior Differences in Bonobos (Pan paniscus)

“…Bonobos have the largest vocal repertoire size of the nonhuman great apes 52 , and modify their communicative signals depending on social context 53,54,55 . In addition, multiple reports in a single bonobo support the notion that bonobos are able to understand spoken English words and sentences 56,57,58 , as well as degraded and computer-generated speech 59 . Thus, it is possible that FOXP2 alleles modulate FOXP2 expression in bonobos and that differential FOXP2 regulation impacts individual variability in bonobo vocal communication.…”

Perception of vocoded speech in domestic dogs

“…The first experiment found that dogs listened significantly longer to their own 16-channel vocoded name than the foil vocoded name when presented with both stimuli in natural speech and vocoded speech. This is similar to the stimulus sets used with Kanzi, a bonobo, and Panzee, a chimpanzee, where interleaved natural speech and vocoded speech were used in forced-choice paradigms (Heimbauer et al 2011 , 2021 ; Lahiff et al 2022 ). As was the case in prior degraded-speech processing studies using nonhuman animals, natural speech was included in this study because the intelligibility of distorted speech (e.g., sine-wave speech or vocoded speech) dramatically increases when people hear the natural version of the sentence prior to hearing the distorted version (Davis et al, 2005 ).…”

“…This is in line with other studies from our lab demonstrating that dogs listen longer to their names than foil names in another version of degraded speech: speech in noise (Mallikarjun et al 2019 ). In all but one of the prior vocoded speech comprehension studies done with non-human animals, the target species were exposed to vocoded speech in a training/orientation period prior to testing their comprehension (Heimbauer et al 2021 ; Lahiff et al 2022 ; Shofner 2014 ; Shofner et al 2018 ). Panzee the chimpanzee was able to spontaneously recognize vocoded speech in a paradigm that also included trials with natural speech, as was tested in Experiment 1 with the dogs.…”

“…Recognition of degraded speech in a nonhuman animal with human language experience has been explored with Kanzi, a language-trained bonobo (Lahiff et al 2022 ), and Panzee, a language-trained chimpanzee (Heimbauer et al 2011 , 2021 ). In similar paradigms, both recognized noise-vocoded tokens of familiar words and correctly mapped them to the corresponding target image or pictogram, presented in sets of three (Lahiff et al 2022 ) or four (Heimbauer et al 2021 ).…”

“…Recognition of degraded speech in a nonhuman animal with human language experience has been explored with Kanzi, a language-trained bonobo (Lahiff et al 2022 ), and Panzee, a language-trained chimpanzee (Heimbauer et al 2011 , 2021 ). In similar paradigms, both recognized noise-vocoded tokens of familiar words and correctly mapped them to the corresponding target image or pictogram, presented in sets of three (Lahiff et al 2022 ) or four (Heimbauer et al 2021 ). In Heimbauer et al ( 2011 ), Panzee achieved above-chance performance (55% accuracy, with chance at 25%) with spontaneous recognition of 7-channel noise-vocoded stimuli during her first exposure to the stimuli; this provides some evidence that language experience contributes to the ability to perceive degraded speech.…”