Enhanced amplitude modulations contribute to the Lombard intelligibility benefit: Evidence from the Nijmegen Corpus of Lombard Speech

Understanding the Lombard Effect for Mandarin: Relation Between Speech Recognition Thresholds and Acoustic Parameters

The present work quantifies the Lombard effect across native speakers of Mandarin Chinese using the Matrix sentence test, which is optimized for precisely assessing speech recognition thresholds (SRTs) in noise. Specifically, we studied the effects of speaker gender, fundamental frequency (F0), formant frequencies (F1 and F2), the duration and rate of voiced segments, and frequency-specific energy redistribution characterized by alpha ratio and speech-weighted signal-to-noise ratio (swSNR) on the recognition of Mandarin in plain and Lombard speech. The Mandarin Chinese matrix test was recorded with plain and Lombard speech from 11 native-Mandarin speakers. SRTs in stationary noise were measured with native-Mandarin, normal-hearing listeners. Results showed that on average, Mandarin Lombard speech was more intelligible than Mandarin plain speech for both female and male speakers, and the Mandarin Lombard gain of female speakers was larger than that of males. In addition, various acoustic analyses involving all speakers showed that (a) only swSNR was significantly correlated with the SRT of the Mandarin plain speech; (b) most acoustic measures were significantly correlated with the SRT of the Mandarin Lombard speech; and (c) alpha ratio and swSNR were significantly correlated with the SRT Lombard gain. In addition, a gender effect was found in the correlational analysis between acoustic parameters and SRT as well as Lombard gain in SRT. The findings highlight the impact of increased high-frequency energy on the observed Lombard gain in Mandarin speech, whereas the changes in individual acoustic parameters (e.g., F0 and F1) appear to play only a minor role.

Speaking in the presence of noise: Consistency of acoustic properties in clear-Lombard speech over time

Individual speakers are often able to modify their speech to facilitate communication in challenging conditions, such as speaking in a noisy environment. Such vocal "enrichments" might include reductions in speech rate or increases in acoustic contrasts. However, it is unclear how consistently speakers enrich their speech over time. This study examined inter-speaker variability in the speech enrichment modifications applied by speakers. The study compared a baseline habitual speaking style to a clear-Lombard style and measured changes in acoustic differences between the two styles over sentence trials. Seventy-eight young adult participants read out sentences in the habitual and clear-Lombard speaking styles. Acoustic differences between speaking styles generally increased nonlinearly over trials, suggesting that speakers require practice before realizing their full speech enrichment potential when speaking clearly in noise with reduced auditory feedback. Using a recent objective intelligibility metric based on glimpses, the study also found that predicted intelligibility increased over trials, highlighting that communicative benefits of the clear-Lombard style are not static. These findings underline the dynamic nature of speaking styles.

Design and Development of a Spanish Hearing Test for Speech in Noise (PAHRE)

BACKGROUND

There are few hearing tests in Spanish that assess speech discrimination in noise in the adult population that take into account the Lombard effect. This study presents the design and development of a Spanish hearing test for speech in noise (Prueba Auditiva de Habla en Ruido en Español (PAHRE) in Spanish). The pattern of the Quick Speech in Noise test was followed when drafting sentences with five key words each grouped in lists of six sentences. It was necessary to take into account the differences between English and Spanish.

METHODS

A total of 61 people (24 men and 37 women) with an average age of 46.9 (range 18-84 years) participated in the study. The work was carried out in two phases. In the first phase, a list of Spanish sentences was drafted and subjected to a familiarity test based on the semantic and syntactic characteristics of the sentences; as a result, a list of sentences was selected for the final test. In the second phase, the selected sentences were recorded with and without the Lombard effect, the equivalence between both lists was analysed, and the test was applied to a first reference population.

RESULTS

The results obtained allow us to affirm that it is representative of the Spanish spoken in its variety in peninsular Spain.

CONCLUSIONS

In addition, these results point to the usefulness of the PAHRE test in assessing speech in noise by maintaining a fixed speech intensity while varying the intensity of the multi-speaker background noise. The incorporation of the Lombard effect in the test shows discrimination differences with the same signal-to-noise ratio compared to the test without the Lombard effect.

Using fuzzy string matching for automated assessment of listener transcripts in speech intelligibility studies

Many studies of speech perception assess the intelligibility of spoken sentence stimuli by means of transcription tasks ('type out what you hear'). The intelligibility of a given stimulus is then often expressed in terms of percentage of words correctly reported from the target sentence. Yet scoring the participants' raw responses for words correctly identified from the target sentence is a time-consuming task, and hence resource-intensive. Moreover, there is no consensus among speech scientists about what specific protocol to use for the human scoring, limiting the reliability of human scores. The present paper evaluates various forms of fuzzy string matching between participants' responses and target sentences, as automated metrics of listener transcript accuracy. We demonstrate that one particular metric, the token sort ratio, is a consistent, highly efficient, and accurate metric for automated assessment of listener transcripts, as evidenced by high correlations with human-generated scores (best correlation: r = 0.940) and a strong relationship to acoustic markers of speech intelligibility. Thus, fuzzy string matching provides a practical tool for assessment of listener transcript accuracy in large-scale speech intelligibility studies. See https://tokensortratio.netlify.app for an online implementation.

Effects of task and language nativeness on the Lombard effect and on its onset and offset timing

This study focuses on the differences in speech sound pressure levels (here, called speech loudness) of Lombard speech (i.e., speech produced in the presence of an energetic masker) associated with different tasks and language nativeness. Vocalizations were produced by native speakers of Japanese with normal hearing and limited English proficiency while performing four tasks: dialog, a competitive game (both communicative), soliloquy, and text passage reading (noncommunicative). Relative to the native language (L1), larger loudness increments were observed in the game and text reading when performed in the second language (L2). Communicative tasks yielded louder vocalizations and larger increments of speech loudness than did noncommunicative tasks regardless of the spoken language. The period in which speakers increased their loudness after the onset of the masker was about fourfold longer than the time in which they decreased their loudness after the offset of the masker. Results suggest that when relying on acoustic signals, speakers use similar vocalization strategies in L1 and L2, and these depend on the complexity of the task, the need for accurate pronunciation, and the presence of a listener. Results also suggest that speakers use different strategies depending on the onset or offset of an energetic masker.

Echolocating bats exhibit differential amplitude compensation for noise interference at a sub-call level

Flexible vocal production control enables sound communication in both favorable and unfavorable conditions. The Lombard effect, which describes a rise in call amplitude with increasing ambient noise, is a widely exploited strategy by vertebrates to cope with interfering noise. In humans, the Lombard effect influences the lexical stress through differential amplitude modulation at a sub-call syllable level, which so far has not been documented in animals. Here, we bridge this knowledge gap with two species of Hipposideros bats, which produce echolocation calls consisting of two functionally well-defined units: the constant-frequency (CF) and frequency-modulated (FM) components. We show that ambient noise induced a strong, but differential, Lombard effect in the CF and FM components of the echolocation calls. We further report that the differential amplitude compensation occurred only in the spectrally overlapping noise conditions, suggesting a functional role in releasing masking. Lastly, we show that both species of bats exhibited a robust Lombard effect in the spectrally non-overlapping noise conditions, which contrasts sharply with the existing evidence. Our data highlight echolocating bats as a potential mammalian model for understanding vocal production control.