Easy-Onset Techniques: Clinical Rationale

Introduction

Stammering (also termed stuttering or dysfluency) is a speech disorder characterised by involuntary disruptions to the flow of speech — repetitions of sounds and syllables, prolongations, and blocks. It affects approximately 1% of adults worldwide and has a significant impact on occupational, social, and psychological wellbeing.

Despite decades of research, stammering has no single pharmacological or surgical cure. The most robustly evidence-based interventions for adults are behavioural speech therapy techniques, particularly the fluency-shaping approach. These techniques train speakers to use a modified speech pattern that, with sufficient practice, reduces or eliminates overt stammering.

This document summarises the clinical rationale for the three core fluency-shaping techniques used in Flowen — easy onset, diaphragmatic breathing support, and prolonged speech — and describes how Flowen implements real-time biofeedback for each.

Fluency shaping vs stammering modification

Clinical management of stammering broadly divides into two approaches:

Flowen's core intervention is fluency-shaping — specifically because this approach is amenable to biofeedback augmentation and daily self-directed practice. The techniques are discrete, teachable, and the behaviours they target (laryngeal onset, breath flow, rate) are measurable in real time.

Easy onset

Theory and rationale

Easy onset (also termed gentle onset or smooth onset) is the application of a gentle, gradual initiation of phonation at the start of an utterance. It directly targets one of the primary mechanisms of stuttering: laryngeal hyperadduction — an involuntary tightening of the vocal folds that creates a "block" before or during speech.

The neural substrates of stuttering are complex and incompletely understood, but converging evidence points to abnormal functioning in the basal ganglia–thalamocortical motor loops responsible for the initiation and timing of speech movements (Chang & Zhu, 2013; Alm, 2004). In people who stammer, the timing and coordination of laryngeal onset is disrupted — the vocal folds either close prematurely (causing a block) or open and close repeatedly (causing repetitions).

Easy onset addresses this by teaching the speaker to initiate voicing with a gradual, soft onset rather than an abrupt one. This reduces the muscular tension at the glottis and gives the speech motor system more time to coordinate the transition from silence to phonation.

Technique

Easy onset is characterised by the following features in production:

• Begin the first syllable of an utterance with a slow, gentle increase in vocal fold vibration
• Air flow should be initiated before voicing begins — the breath "leads" the sound
• The onset should be soft and gradual (like a sigh flowing into speech), not hard and abrupt
• Muscle tension in the jaw, lips, and tongue should be consciously reduced
• The technique is applied especially at the start of utterances and on words beginning with vowels

Flowen biofeedback implementation

Flowen measures the acoustic correlates of vocal onset quality in real time using the device microphone. The key signal is the rate of amplitude rise at the onset of voiced speech — a slow, gradual rise indicates a gentle onset; a steep, abrupt rise indicates a hard onset.

The platform also monitors fundamental frequency (F0) at onset — hard onsets are frequently associated with momentary pitch perturbations as the vocal folds slam shut. Flowen's signal processing pipeline flags both a steep amplitude onset and F0 perturbations as markers for coaching feedback.

Diaphragmatic breathing support

Anatomy of breath support

Speech is produced on exhaled air. The quality and control of that exhalation directly affects fluency. People who stammer frequently exhibit abnormal respiratory patterns during speech — particularly incomplete inhalation before utterances, excess tension in the chest and neck muscles, and premature closure of the airway.

Diaphragmatic breathing (also called abdominal or belly breathing) prioritises use of the diaphragm — the large dome-shaped muscle at the base of the lungs — as the primary driver of inhalation and exhalation, rather than the chest and accessory muscles. This produces a deeper, more controlled breath with a steadier subglottal air pressure, which in turn creates a more stable substrate for phonation.

Key targets in diaphragmatic breathing for speech:

• Full diaphragmatic inhalation before beginning an utterance
• Controlled, sustained exhalation with minimal chest movement
• Avoidance of "speaking on empty" — beginning utterances without adequate breath support
• Relaxation of the abdominal and chest muscles during exhalation

Evidence

Respiratory retraining is a standard component of fluency-shaping programmes, including the Camperdown Programme and the Lidcombe-adjacent approaches. While isolated evidence for breath support is difficult to extract from combined-technique studies, the theoretical rationale is strong: adequate subglottal pressure is necessary for consistent voicing, and people who stammer demonstrably show abnormal respiratory patterns (Zocchi et al, 1990; Peters & Boves, 1988).

Flowen incorporates a breath support phase at the start of each session — a structured breathing exercise that primes the respiratory system before practice. This is based on the evidence that pre-session relaxation and breath training reduce overall tension and improve technique acquisition (Guitar, 2014).

Prolonged speech

Mechanism

Prolonged speech is a technique in which the speaker extends the duration of vowels and the transitions between sounds, effectively reducing overall speech rate. It is one of the oldest and most robustly evaluated fluency-shaping techniques.

The mechanism by which rate reduction improves fluency is multi-factorial:

• Timing: Slowing speech gives the aberrant motor timing loops more time to coordinate correctly
• Continuous phonation: Maintaining airflow and voicing across word boundaries reduces the number of "restart" opportunities where blocks can occur
• Proprioceptive feedback: Slower speech increases the speaker's sensory awareness of their own articulatory movements, supporting self-monitoring
• Reduced cognitive load: At slower rates, the speech planning system has more time to prepare the next segment, reducing the emergency "retrieval" situations that often precipitate blocks

Evidence

Prolonged speech has one of the strongest evidence bases in stuttering therapy. The Camperdown Programme — one of the most extensively evaluated fluency-shaping programmes — is centred on prolonged speech and achieves significant fluency gains in RCT conditions (O'Brian et al, 2003, 2014).

Key findings from the literature:

• O'Brian et al (2014) demonstrated ≥50% reduction in %SS in 70% of adult participants completing Camperdown-based treatment
• Carey et al (2010) showed non-inferiority of internet-delivered Camperdown (including prolonged speech) vs clinician-delivered treatment
• Relapse rates are lower when participants maintain daily self-directed practice of prolonged speech post-discharge (Guitar, 2014)

How Flowen operationalises these techniques

Flowen's biofeedback engine analyses incoming audio in real time — amplitude at 100 frames per second, pitch and tension at 25 analysis blocks per second — and derives the following signals relevant to fluency-shaping technique practice:

These signals are processed by Flowen's audio pipeline — a WebAudio API implementation running client-side — and the results are used to drive three types of feedback:

• Visual waveform feedback: Real-time amplitude display showing the user their onset quality
• Colour-coded tension cue: The voice-tension index bar shifts from blue to amber to orange as it rises toward the threshold that triggers an easy-onset prompt
• AI voice coach: A spoken coaching prompt, triggered by natural pauses in speech or by a high tension reading, giving specific technique feedback during the session rather than only afterward

Evidence summary

References

• Alm, P.A. (2004). Stuttering and the basal ganglia circuits. Journal of Communication Disorders, 37(4), 325–369.
• Bothe, A.K. et al (2006). Stuttering treatment research 1970–2005: I. Systematic review incorporating trial quality assessment of behavioral, cognitive, and related approaches. American Journal of Speech-Language Pathology, 15(4), 321–341.
• Carey, B. et al (2010). Randomized controlled non-inferiority trial of a telehealth treatment for chronic stuttering. Journal of Speech, Language, and Hearing Research, 53(6), 1438–1452.
• Chang, S.E. & Zhu, D.C. (2013). Neural network connectivity differences in children who stutter. Brain, 136(12), 3709–3726.
• Guitar, B. (2014). Stuttering: An Integrated Approach to Its Nature and Treatment (4th ed.). Lippincott Williams & Wilkins.
• Ingham, R.J. et al (2012). Stuttering treatment and brain research in adults: A still unfolding relationship. Journal of Fluency Disorders, 37(4), 218–234.
• O'Brian, S. et al (2003). The Camperdown Program: Outcomes of a new prolonged-speech treatment model. Journal of Speech, Language, and Hearing Research, 46(4), 933–946.
• O'Brian, S. et al (2014). Stuttering severity at 2, 4, and 7 years. Journal of Speech, Language, and Hearing Research, 57(6), 2093–2101.
• Peters, H.F.M. & Boves, L. (1988). Coordination of aerodynamic and phonatory processes in fluent speech utterances of stutterers. Journal of Speech and Hearing Research, 31(3), 352–361.
• RCSLT (2023). Clinical guidance: Fluency disorders. Royal College of Speech and Language Therapists. rcslt.org