How to slow down your speech: do we need a new approach?

Exceptionally fast talking isn’t rare. It’s even got a fancy scientific name:

tachylalia. (It’s Greek, meaning “swift chatter”. )

I don’t like the term:

  • It’s pointless jargon for excessively rapid speech.
  • It sounds like a “disorder”, which is not always the case.

Why another article on slowing down speech rate?

Three reasons:

  • I’ve been working on it with several clients recently.
  • I’m interested in it. Rate control is complex and hard to achieve. I love a good challenge!
  • The research base is limited. I wanted to share some of the research and clinical reasoning I’ve used in practice.

I’ve written about traditional rate control techniques before. But they don’t work for some clients, so I’ve being trying new things.

What causes exceptionally fast talking?

  • Sometimes, fast talking is caused by a brain injury. For example, we’ve known for a long time that some people with traumatic brain injuries, hypokinetic dysarthria, cerebella dysarthria and progressive supernuclear palsy speak really quickly (e.g. Darley et al., 1969, Verruips et al., 1998, Hansen & Metter, 1980; and Hardy, 1967).
  • Some people who “clutter” have abnormally rapid or irregular speech (or both) (St Louis et et., 2007). You can read more about cluttering here.
  • Other people just speak really quickly for no apparent reason (e.g. Bakker et al., 2011).

How fast is too fast?

Some people have proposed arbitrary syllable-a-minute-counts. But I’m not aware of any research that supports this approach.

I like Bakker’s definition:

“Exceptionally rapid speech is speech that draws attention to itself as being faster than normal and that potentially reduces intelligibility because it occurs faster than some listeners can comprehend.”

When I’m contacted by a client seeking help to slow their speech rate, it’s usually because they are self-conscious about how quickly they speak, and haven’t been able to slow down alone.

Why slow down?

Many speech pathologists (including me) consider speaking at a slower rate to be the best way to improve how easy you are for others to understand (e.g. Yorkston et al., 1992, 1999).

Speaking at a slower rate can improve speech by helping the speaker to get and keep voluntary control over speech output. It can:

  • improve coordination of your breath, voice, jaw, tongue, teeth, and lips (e.g. Kent et al., 1975);
  • improve your breathing patterns (e.g. Yorkston et al., 1990); and
  • give the listener added processing time to understand what you are saying (Yorkston et al., 1999).

The last point is crucial. Some fast talkers with good articulation skills don’t recognise that their rate is simply too fast for their listeners to process.

How is fast talking traditionally treated? Flexible and rigid approaches

Clients with very fast speech may have had years of friends, family, workmates and sometimes even complete strangers advising them to “just slow down”. If only it were so easy!

Some researchers divide traditional clinical procedures for slowing speech rate into flexible and rigid approaches:

  • Flexible approaches: I like these because they don’t require fast talkers to learn an artificial “robotic” pattern. They include:
    • instructions to slow down while copying the speech pathologist who “models” a slow rate with natural intonation (e.g. Caliguiri, 1989);
    • rhythmic cueing: the speech pathologist signals the slower rate by pointing to words of a written passage in a rhythmic fashion (e.g. Yorkson & Beukelman, 1981);
    • bio-feedback systems: where the fast talker uses visual feedback (e.g. from a oscilloscope) to monitor his/her rate (e.g. Berry & Goshorn); and
    • prolonged speech: the speaker learns a new speaking pattern usually involving longer vowels, soft onsets and continuous voicing. Prolonged speech training is very common in evidence-based stuttering treatments like the Camperdown Program.
  • Rigid approaches: I don’t like these unless flexible approaches have been exhausted without results, or if the the client has a very severe rate control problem. They result in robotic, unnatural or one word at a time speech patterns, which are rarely acceptable in professional and social contexts. They include:
    • pacing boards: the individual points to coloured slots on a board as he/she says each word, i.e. one word to a dot. They have been shown to help some clients with Parkinson’s Disease with severe palilalia (Helm, 1979);
    • alphabet supplementation: the speaker points to the first letter of each word spoken (e.g. Yokston, 1999). This can help some clients with dysarthrias (e.g. Crow & Enderby, 1989);
    • hand or finger tapping: the speaker taps his/her hand or finger as he/she speaks in a syllable-by-syllable pattern (e.g. Duffy, 1995);
    • metronome treatment: the speakers talks to a beat, sometimes at around 90 beats a minute; and
    • delayed auditory feedback: this delays the rate at which the speaker’s speech is fed back to him/her through headphones (e.g. Hanson & Metter). I’ve never used it, probably because it requires expensive equipment and has a very patchy evidence base in stuttering therapy.

Is there a better way?

In clinical practice, some people respond quickly to simple cues to slow down (e.g. Turner et al., 1995). Other people can be very resistant to the therapies above (e.g. Yorkston, 1999). Sometimes, you can achieve success with short sentences and reading, but the results don’t “transfer” across to real world conversations. For these reasons, I’ve become increasingly dissatisfied with traditional treatments.

Here are some factors that inform my clinical approach to helping people who speak too quickly:

  • Insight into the problem is a good first step: Recently, I have been using identity mapping approaches to help clients understand how they are perceived by others. This can involve activities like recording and playing back audio and video, agreeing cues to signal the need to slow down, analysing the speech rate of personal heroes and public speakers, negative and positive practice, and creating a professional “avatar” of the perfect speaker identity (from the client’s perspective) to work toward. You can read more about this approach and the evidence supporting it here.
  • Acceptance of the new rate and speech pattern is essential. Many clients will reject techniques that leave them sounding unnatural or “phony”. I don’t blame them. Just as with adult stuttering treatments, it’s important to reassure clients that you will not leave them speaking with an unusual pattern “out in the real world”, even if you are practising a pattern in the clinic as a stepping stone to effecting permanent change.
  • Motivation to practice are key for rate control treatments to work. Clients who are determined to slow their rate – e.g. to become better speakers or to get a promotion – might buy into therapy more than clients who are attending therapy under duress (e.g. because their partner or boss told them to do it). For some people, speaking too quickly has no real effect on their work and life goals. If this is the case, low motivation to change may be an obstacle to treatment.
  • Principles of motor learning. As you may have noticed, the evidence base for many of the traditional rate control techniques is pretty dated. Rate control is a motor skill. Therefore, therapies should ideally be based on modern principles of motor learning. This may mean multiple sessions that provide the learner with hundreds (or even thousands) of trials within each session (e.g. McNeil et al., 1997). You can read more about principles of motor learning here.

These ideas underpin what some researchers have termed “backdoor approaches” to rate control. These are therapy approaches that have been found to result in a slowed down speech rate as a byproduct of working on something else. Here are some examples of back-door approaches I’ve used in this clinic:

  • LSVT LOUD: This treatment was originally designed (and is still primarily used) to treat the people with Parkinson’s Disease. It’s based on principles of motor learning, and has one global target: increased volume. It has been shown that targeting volume with this treatment reduces average speaking rate (e,g, Ramig et al., 1995). You can read more about LSVT here.
  • Naturalness treatments: These treatments target speech naturalness instead of rate reduction. This can be done by lengthening syllable duration, working on emphatic stress, breath patterns or phrasing. One technique I’ve used in the clinic is to have the client break speech into “thought units” (roughly analogous to written sentences or clauses), pausing between each, and stressing one or two words for each thought unit, e.g. “Today, I going to talk about five key events”). I’ve found this to be particularly useful when working on public speaking skills for professionals with rate control problems.
  • Feedback: Giving naturalness ratings to clients about their speech has been shown to improve sentence intelligibility for some speakers (e.g. Ice & Rosenberg, 1973). The latest research tells us that feedback should be based of principles of motor learning: e.g. feedback should be random, infrequent and delayed.
  • Focus on comprehensibility: Comprehensibility is all about how easy a person is for another person to understand. This is a “top down approach” – a bit like identity mapping – where a speaker is taught to: (a) tell listeners that he/she speaks too quickly and wants to know if he/she is understood; (b) attend to social and environmental cues (e.g. body language suggesting listeners have switched off); (c) check in with listeners as to whether he/she is understood; and (d) ensure listeners are listening before speaking (e.g. Vogel & Miller, 1991).

Clinical bottom line

Exceptionally rapid speech is a real problem for some people. It can get in the way of achieving work, social and life goals. Traditional treatment approaches – flexible and rigid – work for some. But, for others, these approaches fail to achieve permanent change. For these clients, “backdoor approaches” based on principles of motor learning, identity mapping, natural sounding outcomes and improved comprehensibility are well worth trialling. Speech pathologists who work in this area should keep a close eye on new peer-reviewed research on rate control. In the meantime, we should share what works and what doesn’t work in the clinic and test our approaches by measuring treatment outcomes objectively.

Related articles:

Principal sources:

  1. Marshall, R.C., Karow, C.M. (2002). Retrospective Examination of Failed Rate-Control Intervention. American Journal of Speech-Language Pathology, 11, 1: 3-16. I like this study a lot. The researchers tried a whole host of methods on a client and none of them worked! Relatively few studies of ineffective speech pathology treatments get published. This is called “publication bias” and results in speech pathologists reinventing the wheel and trialling things that don’t work because we don’t know about studies showing they don’t work!
  2. Bakker, K., Myers, F.L., Raphael, L.J., and St Louis, K.O. (2011). A preliminary comparison of speech rate, self-evaluation, and disfluency of people who speak exceptionally fast, clutter, or speak normally, in Scaler et al., Cluttering, pages 46-65, Taylor & Francis. This is also an interesting study. The researchers were unable to find a clinically reliable way of distinguishing people who speak exceptionally fast from people who clutter, with both groups able to slow down in highly structured sentence-level activities.



Hi there, I’m David Kinnane.

Principal Speech Pathologist, Banter Speech & Language

Our talented team of certified practising speech pathologists provide unhurried, personalised and evidence-based speech pathology care to children and adults in the Inner West of Sydney and beyond, both in our clinic and via telehealth.

David Kinnane
Speech-Language Pathologist. Lawyer. Father. Reader. Writer. Speaker.

Leave a Reply

Your email address will not be published.

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Share This

Copy Link to Clipboard