Fact checked byKristen Dowd

Read more

September 16, 2024
2 min read
Save

Speech features change during onset, peak of COPD exacerbation

Fact checked byKristen Dowd
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Key takeaways:

  • Minimum pitch in the morning was significantly higher at exacerbation onset and peak vs. baseline.
  • In the evening, jitter significantly differed between baseline and the exacerbation time points.

Patients with COPD experience changes in voice pitch and jitter during the onset and peak of an exacerbation, according to a poster presented at the European Respiratory Society International Congress.

Loes van Bemmel

“Our findings pave the way for future tools to help people manage their own exacerbations with speech applications,” Loes van Bemmel, MSc, PhD student and researcher in the department of respiratory medicine at Maastricht University Medical Center in the Netherlands, told Healio. “Hopefully our findings help lessen the burden on everyday clinicians in the future, and for now convince people of the importance and potential use of speech analysis.”

Close up of man recording his voice with a smart phone.
Patients with COPD experience changes in voice pitch and jitter during the onset and peak of an exacerbation, according to a poster presentation. Image: Adobe Stock

Using the TACTICAS smartphone app, van Bemmel and colleagues assessed speech recordings of 28 patients with COPD taken three times a day for 12 weeks to examine differences in speech at onset and peak of symptoms of a COPD exacerbation vs. speech at baseline.

Patients filled out the EXACT questionnaire once a day, so researchers knew when they experienced an exacerbation (score 9 points above baseline for 3 days or 12 points above baseline for 2 days).

According to a press release from the European Respiratory Society, recordings captured patients “saying ‘ahh’ for as long as they could manage with one breath, then either reading a short paragraph of a story or answering a question.”

The Praat computer program allowed researchers to extract 25 speech features, such as pitch, voice breaks and jitter from sustained vowel /a/ in patients’ morning and evening recordings.

By the 4-week mark, 11 patients (mean age, 68.6 years; 36.4% women) had at least one exacerbation, and by the 8-week mark, the study cohort reported 16 COPD exacerbations.

In the morning recordings, researchers observed a significant difference between baseline and exacerbation onset in terms of minimum pitch (73.84 Hz vs. 101.48 Hz; P < .05) and number of voice breaks (11.75 vs. 5.38; P < .05). Similarly, minimum pitch in the morning was significantly higher at exacerbation peak vs. baseline (106.59 Hz vs. 73.84 Hz; P < .01) and the number of voice breaks was significantly lower (4.6 vs. 11.75; P < .05).

In the evening recordings, the speech feature that significantly differed between baseline and the exacerbation time points was jitter (breathy or hoarse sounding voice). Local absolute jitter was significantly higher at exacerbation onset and peak vs. baseline (118.39 µs and 104.23 µs vs. 100.07 µs; P < .05). Additionally, jitter measured as five-point period perturbation quotient was significantly elevated at onset of an exacerbation vs. baseline (0.94% vs. 0.62%; P < .05) and at peak of an exacerbation vs. baseline (0.84% vs. 0.62%; P < .05).

“Even though we hoped for it, we were surprised to see that the speech characteristics already change significantly the first day of the exacerbation,” van Bemmel told Healio. “We expected the speech to be different during the peak of symptoms, but finding significant difference on the onset day is very nice.

“In future studies we will validate our findings with more data and design algorithms to predict exacerbations by speech as well,” van Bemmel said. “Furthermore, we will design an app specifically for people with COPD to also employ the algorithm.”

References: