Measuring speaking rate: how do objective measurements correlate with audio-perceptual ratings?

OBJECTIVE: Although speaking rate is central for many speech disorders, no consensus exists regarding the measurement of this feature. The purpose of this study was to examine the correlations between perceptual listener evaluations and various measures of speaking rate. Furthermore, the study investigated the relationship between speaking rate and articulation rate and how pauses in speech affect the perceived tempo. METHOD: Nine healthy females were selected to produce stimuli representing three habitual speech tempi during semi-spontaneous speech: slow (n = 3), neutral (n = 3) and fast (n = 3). Speaking rate was analyzed both by manual calculation and through automatic detection by a script to the computer-based program Praat. Thirty untrained male and female listeners evaluated the recordings with regard to speech tempo on visual analogue scales from very slow to very fast. RESULTS: Large, significant correlations (Pearson's r) were found between all objective measures of speaking rate and perceptual listener evaluations. Words/minute showed the largest correlation (.91), followed by syllables/second (.89), while articulation rate (pauses excluded) as automatically measured by the script, showed the smallest correlation (.69). Possible explanations for the findings are discussed. CONCLUSION: Untrained listeners' evaluation of speech tempo in normal subjects correlated strongly with objective measurements. The results both support the use of auditive-perceptual evaluation of tempo and the use of automatic script analysis for clinical use. Speaking rate (pauses included) showed better consistency with perceptual listener evaluations than articulation rate.

Vocal Outcome After Cordectomy by Transoral CO2 Laser Microsurgery in Patients With Laryngeal Intraepithelial Neoplasia and Non-neoplastic Lesions.

OBJECTIVES: This study investigates vocal outcome after cordectomy by transoral CO2 laser microsurgery (TLM-cordectomy) in patients with laryngeal intra-epithelial neoplasia (LIN) or non-neoplastic lesions (NNL), for improved individual patient advice and potential adjustment of national treatment strategy by which patients suspected to have glottic LIN or T1a cancer are offered TLM-cordectomy, without prior biopsy. STUDY DESIGN: Prospective, longitudinal, quasi-experimental time series. METHODS: Consecutively included patients (n = 155) with LIN (n = 84) or NNL (n = 71) who underwent voice assessments before and after TLM-cordectomy. The multi-dimensional voice assessment protocol comprised voice and speech range profiles, aerodynamics, acoustic analysis, self-evaluated voice handicap, and perceptual auditory voice ratings. RESULTS: Median follow-up time was 195 (range 50-1121) days for patients with LIN and 193 (range 69-1294) days for patients with NNL. Statistically significant changes, LIN: in voice handicap index (VHI) and breathiness after TLM-cordectomy. Statistically significant changes, NNL: voice range profile (voice range area, intensity range, and frequency range) and VHI after TLM-cordectomy. All group-wise changes were to less disordered voices. Previous smokers had the largest decreases in VHI and breathiness. Patients with baseline VHI scores >65 had smaller increases in VHI, however 13-19% of the patients had increases in VHI above the clinically relevant threshold after TLM-cordectomy. CONCLUSION: Overall, TLM-cordectomy in patients with LIN and NNL improved vocal outcome and our study thus supports the current Danish treatment strategy and improves the basis for proper patient advice. Multi-dimensional voice assessment is suggested preoperatively and six-nine months postoperatively, with focus on individual vocal differences and voice demands.

Voice Therapy Outcome: A Comparison Between Imitation Model Voice Therapy and Verbal Instructions Model Voice Therapy.

OBJECTIVES: The aim was to compare voice outcomes over time in patients treated with Imitation Model voice therapy and Verbal Instructions Model voice therapy. METHODS: A prospective clinical trial was performed with 56 consecutive patients diagnosed with a primary or secondary functional voice disorder. Thirty-one patients were included for voice therapy following the Imitation Model and 25 patients for the Verbal Instructions Model. Assessments included a self-rated Voice Handicap Index, self-perceived hoarseness and vocal fatigue, perceptual voice evaluation by a Speech Language Pathologist, and maximum Voice Range Profiles. All assessments were completed before therapy, at end of therapy, at six-months posttherapy follow-up and 12-months posttherapy follow-up. For maximum Voice Range Profiles group differences were also compared for effects from end-of-treatment to follow-up assessments. Linear mixed models were used for analysis. RESULTS: Comparison between treatment groups showed significantly larger long-term improvement from the baseline, for verbal instructions model as compared to imitation model for Voice Handicap Index total, and also the physical and emotional subscales, while there was no difference between groups for the functional subscale. Also, voice quality improved more after verbal instructions model, as compared to imitation model, at end of therapy. Results from self-rated hoarseness and vocal fatigue showed no difference between groups. There was no difference between treatment groups in the change of maximum Voice Range Profile from end-of-treatment to follow-up assessments. CONCLUSIONS: The study showed that both Imitation Model voice therapy and Verbal Instructions Model voice therapy improved voice function. Compared to Imitation Model, the Verbal Instructions model showed larger long-term effect on physical and emotional aspects of voice function in everyday life. The two approaches for voice therapy might have different impacts on patients' learning.

Effects of Speaking Rate on Breathing and Voice Behavior.

OBJECTIVES: The objective of this study was to investigate the effects of speaking rate (habitual and fast) and speech task (reading and spontaneous speech) on seven dependent variables: Breath group size (in syllables), Breath group duration (in seconds), Lung volume at breath group initiation, Lung volume at breath group termination, Lung volume excursion for each breath group (in % vital capacity), Lung volume excursion per syllable (in % vital capacity) and mean speaking Fundamental frequency (fO). METHODS: Ten women and seven men were included as subjects. Lung volume and breathing behaviors were measured by respiratory inductance plethysmography and fO was measured from audio recordings by the Praat software. Statistical significance was tested by analysis of variance. RESULTS: For both reading and spontaneous speech, the group increased mean breath group size and breath group duration significantly in the fast speaking rate condition. The group significantly decreased lung volume excursion per syllable in fast speech. Females also showed a significant increase of fO in fast speech. The lung volume levels for initiation and termination of breath groups, as well as lung volume excursions in % vital capacity, showed great individual variations and no significant effects of rate. Significant effects of speech task were found for breath group size and lung volume excursion per syllable, where reading induced more syllables produced per breath group and less % VC spend per syllable as compared to spontaneous speech. Interaction effects showed that the increases in breath group size and breath group duration associated with fast rate were significantly larger in reading than in spontaneous speech. CONCLUSION: Our data from 17 vocally untrained, healthy subjects showed great individual variations but still significant group effects regarding increased speaking rate, where the subjects seemed to spend less air per syllable and inhaled less often as a consequence of greater breath group sizes in fast speech. Subjects showed greater changes in breath group patterns as a consequence of fast speech in reading than in spontaneous speech, indicating that effects of speaking rate are dependent on the speech task.

Unintended imitation of fundamental frequency in female speakers.

Objective: Former studies have shown that humans unintentionally adopt speech characteristics from their conversation partners, a phenomenon known as speech convergence. This study investigated imitation of fundamental frequency (Fo) in nine healthy females in a repetition task.Method: The design included two conditions; loud reading (baseline) and repetition after a model talker. The model talker was a female with a high-pitched voice and stimuli included both words and nonwords. All productions were analyzed with regard to mean and SD of Fo and compared between conditions and word types.Results: The participants showed different patterns regarding the effect on Fo, thus indicating that some individuals are more likely to change their speaking Fo in a repetition task than others. Group results showed somewhat but not significantly higher mean Fo in the repetition task as compared to baseline (reading). Moreover, nonwords were produced with a significantly higher mean Fo than real words, in both reading and repetition conditions. No interaction effect between condition and word type was found.

Mean fundamental frequency in connected speech and sustained vowel with and without a sentence-frame.

Introduction: Sustained vowel phonation is a common patient task in the assessment of voice disorders. Free from articulatory influence, it allows for acoustic voice analyses unfeasible for connected speech. Previous research indicates that the fundamental frequency (fo) of a self-chosen vowel is not necessarily representative to mean fo of the same subject's connected speech.Objective: The purpose was to examine mean fo in text reading and to compare it with sustained vowel phonation with and without a sentence-frame. It was hypothesized, that a short phrase of context embedding the prolonged vowel would affect mean fo of the vowel to be closer to that of text reading.Method: Thirteen vocally healthy women, age 22-27 was audio recorded during three tasks: reading of a standard text, producing an isolated, sustained /a/ and repeating three times a frame sentence with an imbedded /a/ and the third time to sustain the /a/. Mean fo of the three conditions was analyzed by means of the Praat computer program.Results: The isolated vowel condition differed significantly from the mean fo of the sentence-framed one, the latter being closest to mean fo of the reading task. The isolated vowel condition also differed significantly from text reading. The difference in mean fo between sentence-framed vowel and text reading was not significant.Conclusion: Data support that the use of a sentence-frame is helpful to get a client's fo elicited from sustained vowel production in voice assessments more representative to that of text reading as compared to an isolated vowel.

The Impact of Post-thyroidectomy Paresis on Quality of Life in Patients with Nodular Thyroid Disease.

OBJECTIVE: To investigate the impact of postoperative paresis on disease-specific quality of life (DSQoL) after thyroidectomy in patients with benign nodular thyroid disease. STUDY DESIGN: Observational study. SETTING: University hospital. SUBJECTS AND METHODS: Patients were evaluated before and 3 weeks and 6 months after surgery in an individual prospective cohort study using videolaryngostroboscopy (VLS), voice range profile, voice handicap index (VHI), multidimensional voice program, maximum phonation time (MPT), and auditory perceptual evaluation. Changes in DSQoL were assessed by the Thyroid-specific Patient-Reported Outcome measure. Cohen's effect size was used to evaluate changes. RESULTS: Sixty-two patients were included, 55 of whom completed all examinations. Three weeks after surgery, a blinded VLS examination showed signs of paresis of either the recurrent laryngeal nerve or the external branch of the superior laryngeal nerve (RLN/EBSLN) in 13 patients (24%). A paresis corresponded to a 12 ± 28 point increase in VHI (P = .002) and was associated with a significant 4.3 ± 7.5 semitone decrease in the maximum fundamental frequency (P < .001) and a 5.3 ± 8.2 dB reduction in maximum intensity. Further, it was associated with a 4.5 ± 11.2 second reduction in MPT (P = .001) and an increase of 0.40 ± 1.19 in grade, 0.42 ± 1.41 in roughness, and 0.36 ± 1.11 in breathiness. Signs of postoperative RLN/EBSLN paresis correlated with an 11.0-point (P = .02) poorer improvement in goiter symptoms at both 3 weeks and 6 months after surgery. CONCLUSION: Signs of RLN/EBSLN paresis after thyroidectomy were associated with less pronounced improvement in goiter symptoms in patients with thyroid nodular disease. However, thyroidectomy was associated with an overall improved DSQoL by 6 months after surgery.

Auditory-Perceptual Evaluation of Dysphonia: A Comparison Between Narrow and Broad Terminology Systems.

OBJECTIVE: In auditory-perceptual voice analysis, a multiparameter approach and a more reductionist approach may be compared with narrow and broad phonetic transcription and used interchangeably, depending on the purpose. The aim of this study was to investigate the perspectives of a translation of the terminology used in the multiparameter Danish Dysphonia Assessment (DDA) approach into the five-parameter GRBAS system. METHODS: Voice samples illustrating type and grade of the voice qualities included in DDA were rated by five speech language pathologists using the GRBAS system with the aim of estimating inter- and intrarater reliability. The same samples were then rated using the DDA terminology. RESULTS: Both inter- and intrarater reliability were found to be very high for the GRBAS parameters grade, rough, and breathy, but somewhat lower for asthenic and strained. Further, strong and clear associations were found between the DDA and GRBAS rating for grade, rough, breathy, and strained, whereas the relation between DDA ratings and asthenic was weaker and less clear. CONCLUSION: The data strongly support that the DDA system can be translated into the GRBAS system for auditory-perceptual voice analysis. The consensus discussion prior to the listening test is believed to have contributed to the high degree of inter- and intrarater reliability. We suggest for future use of the GRBAS system that rater reliability for asthenic and strained can increase, if these parameters are defined as behavioral terms and antagonists, reflecting muscular hypo- and hyperfunction.

Cognitive Load in Voice Therapy Carry-Over Exercises.

Purpose: The cognitive load generated by online speech production may vary with the nature of the speech task. This article examines 3 speech tasks used in voice therapy carry-over exercises, in which a patient is required to adopt and automatize new voice behaviors, ultimately in daily spontaneous communication. Method: Twelve subjects produced speech in 3 conditions: rote speech (weekdays), sentences in a set form, and semispontaneous speech. Subjects simultaneously performed a secondary visual discrimination task for which response times were measured. On completion of each speech task, subjects rated their experience on a questionnaire. Results: Response times from the secondary, visual task were found to be shortest for the rote speech, longer for the semispontaneous speech, and longest for the sentences within the set framework. Principal components derived from the subjective ratings were found to be linked to response times on the secondary visual task. Acoustic measures reflecting fundamental frequency distribution and vocal fold compression varied across the speech tasks. Conclusions: The results indicate that consideration should be given to the selection of speech tasks during the process leading to automation of revised speech behavior and that self-reports may be a reliable index of cognitive load.

Facilitating behavioral learning and habit change in voice therapy--theoretic premises and practical strategies.

A typical goal of voice therapy is a behavioral change in the patient's everyday speech. The SLP's plan for voice therapy should therefore optimally include strategies for automatization. The aim of the present study was to identify and describe factors that promote behavioral learning and habit change in voice behavior and have the potential to affect patient compliance and thus therapy outcome. Research literature from the areas of motor and behavioral learning, habit formation, and habit change was consulted. Also, specific elements from personal experience of clinical voice therapy are described and discussed from a learning theory perspective. Nine factors that seem to be relevant to facilitate behavioral learning and habit change in voice therapy are presented, together with related practical strategies and theoretical underpinnings. These are: 1) Cue-altering; 2) Attention exercises; 3) Repetition; 4) Cognitive activation; 5) Negative practice; 6) Inhibition through interruption; 7) Decomposing complex behavior; 8) The 'each time-every time' principle; and 9) Successive implementation of automaticity.

Reflections on clinical expertise and silent know-how in voice therapy.

The concept of 'clinical expertise' is described as a part of evidence-based practice (EBP) together with 'external scientific evidence' and 'patient values and perspectives'. However, clinical expertise in the management of voice disorders has not been described or discussed in much detail. The expertise seems to consist partly of silent know-how that, from the outside, may seem improperly related to the personality of the speech-language pathologist or exclusively dependent on the number of years in the field. In this paper, it is suggested that clinical expertise in voice therapy consists of specific skills that can be explicitly described and trained. These skills are discussed together with educational aspects that contribute to the development of clinical expertise. The skills are also discussed from the perspectives of the past, present, and future.

Impact of syllable stress and phonetic context on the distribution of intermittent aphonia.

The occurrence of intermittent aphonia, perceived as sudden interruptions of voicing in connected speech, often reflects high stiffness of the vocal fold mucosa as part of a voice disorder. This retrospective study aimed to investigate the overarching hypothesis that the aphonic instances in voices with intermittent aphonia are not totally randomly appearing, but related to syllable stress and phonetic context. Recordings of 31 dysphonic patients with intermittent aphonia reading a standard text were analyzed perceptually. All vowels of the text were labelled and categorized with regard to syllable stress and character of the phoneme preceding the vowel. The occurrence of aphonic instances within each syllable category was analyzed. Four different hypotheses were formulated and analyzed by the non-parametric Wilcoxon's signed-ranks test. The results showed a significantly higher occurrence of aphonic instances in unstressed syllables as opposed to stressed, in vowels following an unvoiced phoneme as opposed to a voiced, and in vowels following two or more unvoiced phonemes as opposed to one unvoiced phoneme. No significant difference was found between vowels following aspirated stops [p], [t], [k] as opposed to unaspirated stops [b], [d], [g]. The findings support the theory that both physiological and functional aspects may contribute to the phenomenon of intermittent aphonia.

Effects of consensus training on the reliability of auditory perceptual ratings of voice quality.

OBJECTIVES/HYPOTHESIS: This study investigates the effect of consensus training of listeners on intrarater and interrater reliability and agreement of perceptual voice analysis. The use of such training, including a reference voice sample, could be assumed to make the internal standards held in memory common and more robust, which is of great importance to reduce the variability of auditory perceptual ratings. STUDY DESIGN: A prospective design with testing before and after training. METHODS: Thirteen students of audiologopedics served as listening subjects. The ratings were made using a multidimensional protocol with four-point equal-appearing interval scales. The stimuli consisted of text reading by authentic dysphonic patients. The consensus training for each perceptual voice parameter included (1) definition, (2) underlying physiology, (3) presentation of carefully selected sound examples representing the parameter in three different grades followed by group discussions of perceived characteristics, and (4) practical exercises including imitation to make use of the listeners' proprioception. RESULTS: Intrarater reliability and agreement showed a marked improvement for intermittent aphonia but not for vocal fry. Interrater reliability was high for most parameters before training with a slight increase after training. Interrater agreement showed marked increases for most voice quality parameters as a result of the training. CONCLUSION: The results support the recommendation of specific consensus training, including use of a reference voice sample material, to calibrate, equalize, and stabilize the internal standards held in memory by the listeners.