5 seconds of silence
Measures the background sound before the voice recordings begin.
VOICE OUTCOME RECORD
Clinical acoustic voice analysis before and after intervention.
VoiceVitals uses automated Praat analysis to provide acoustic measures for longitudinal clinical review. Reliable interpretation depends on recording quality and forms one part of a wider voice assessment.
Designed as a practical supplement to the voice clinic, it guides recording before and after intervention and compares each patient with their own baseline rather than presenting a diagnostic score.
Current testing status: staff and healthy-volunteer feasibility testing only. Results must not be used for diagnosis or treatment decisions.
WHAT THE PATIENT RECORDS
A short, guided sequence with one instruction shown at a time.
Three comfortable “ah” sounds
An on-screen timer shows how long to hold each attempt.
A fixed Rainbow Passage segment
The same short, standardised text is displayed at baseline and follow-up.
Results sent to the clinician
Praat analysis runs automatically; the audio files are then deleted.
FOR CLINICIANS
A simple addition to the voice-clinic pathway.
Use VoiceVitals to collect comparable acoustic voice samples before and after intervention, using recording tasks appropriate to the clinical pathway. It supplements—not replaces—the clinical history, patient-reported outcomes, perceptual assessment and laryngeal examination.
Create a coded record
Select the procedure or clinical pathway without entering patient-identifiable information.
Generate the patient link
Choose baseline or follow-up, then copy the secure, time-limited assessment link.
Patient completes the tasks
The link guides the patient through the room check, three sustained vowels and a fixed Rainbow Passage segment.
Review and export
Results appear automatically in the clinician record for comparison and PDF export.
RECORDING REQUIREMENTS
What is needed for a reliable acoustic comparison.
Recording quality affects acoustic measurements. These practical requirements are based on published guidance for clinical voice recording and are shown again to the patient before the assessment begins. Read the Praat primer and ASHA instrumental protocol.
A suitable microphone
A wired headset or USB microphone is preferred where available. A device microphone can still be used; allow browser microphone access and do not cover it.
Consistent positioning
Keep the microphone 5–10 cm from the corner of the mouth, slightly to one side to reduce breath noise. Keep it fixed throughout the recording.
A quiet environment
Close doors and windows and switch off avoidable noise such as television, music and fans. VoiceVitals checks five seconds of room noise first.
The same setup at follow-up
Where possible, repeat the assessment using the same device, microphone, room and mouth-to-microphone distance.
GUIDED RECORDING
A clear recording pathway for comparable assessments.
01Prepare
Confirm the recording conditions
The patient sits comfortably and positions the microphone 5–10 cm from the corner of the mouth, slightly off-axis and away from direct breath flow.
- Same microphone where possible
- Quiet room
- Fixed microphone position
02Calibrate
Record five seconds of room noise
The browser measures the device-relative noise floor. Each voice sample is then assessed against that background using signal-to-noise ratio.
- 5-second sample
- Noise floor in dBFS
- Take-level SNR check
03Record
Complete the guided voice tasks
Three comfortable sustained /a/ vowels are followed by a fixed segment of the Rainbow Passage. The patient sees one instruction at a time.
- Three sustained /a/ takes
- Fixed Rainbow Passage segment
- Automatic quality feedback
04Compare
Review change from the patient's baseline
Once the patient submits, the calculated measurements appear automatically in the clinician record and can be exported as a PDF. Audio is not retained after processing.
- Own-baseline comparison
- Quality log
- PDF report
REPORTED PARAMETERS
Praat-derived acoustic measures for longitudinal review.
ParameterUnitTaskDefinition
Mean F0HzSustained /a/Average fundamental frequency, reported with standard deviation.
Jitter (local)%Sustained /a/Cycle-to-cycle frequency perturbation.
Shimmer (local)%Sustained /a/Cycle-to-cycle amplitude perturbation.
HNRdBSustained /a/Praat harmonics-to-noise ratio across the stable voiced segment.
NHRratioSustained /a/Noise-to-harmonic ratio derived from Praat HNR for consensus-aligned reporting.
CPPSdBRainbow PassageSupplementary smoothed cepstral peak prominence from a fixed connected-speech segment.
MPTsSustained /a/Longest valid phonation from three attempts.
Browser microphones are not calibrated sound-level meters. Room background noise is therefore reported in dBFS and used to calculate signal-to-noise ratio, not physical dB(A). VoiceVitals accepts recordings from 15 dB so smartphone users can complete the pathway; recordings below the 30 dB target for perturbation measures are clearly marked as limited quality. An SNR of 42 dB or higher is preferred.
CLINICAL CONTEXT
One part of a multidimensional voice assessment.
The 2023 ELS–UEP consensus describes voice-quality assessment as multidimensional. VoiceVitals provides the acoustic and maximum-phonation component only; findings should be considered alongside history, VHI/VHI-10, clinician perceptual assessment and videolaryngostroboscopy. Read the consensus paper.
PATIENT-REPORTED
VHI or VHI-10, completed separately as part of the clinical pathway.
PERCEPTUAL
Clinician auditory-perceptual rating using GRBAS or the locally agreed method.
LARYNGEAL
Laryngoscopy or stroboscopy findings documented within the clinical record.
ACOUSTIC — VOICEVITALS
Standardised vowel, connected speech and maximum phonation measurements over time.