For speech and language professionals
You cannot recommend what you cannot inspect. Everything in the app is on this page and on the routine page, readable without starting a session.
Last updated 16 September 2026
What this is
A free, browser-based set of speech practice exercises. No account, no sign-up, no app store, no microphone, no data collection, no cost, no upsell. It opens and it runs, including offline.
It is built for a general adult audience who want to warm up or practise speaking aloud: before a meeting, before teaching, or as a daily habit. It is explicitly not a treatment, an assessment, or a substitute for therapy, and the app says so on the home screen, in the completion screen, and in the terms.
What it does not claim
We would rather state this plainly than have you find it.
- The routine has not been clinically validated. Not as a whole, not at any of its durations, and not for any population. It is a sequence assembled from generally described techniques, not a protocol with an evidence base of its own.
- It is not a treatment for stuttering. There is no fluency shaping instruction, no stuttering modification instruction, no delayed auditory feedback, and no desensitisation work. We have removed the word "fluency" from our own marketing for this reason.
- It is not a treatment for dysarthria, apraxia or aphasia, and it is not LSVT LOUD, SPEAK OUT!, or any other named programme, nor a substitute for one.
- It does not strengthen anything. There is no resistance, no effort target, and no strength claim. Every instruction says comfortable effort.
- It does not measure. There is no scoring, no rating and no feedback on performance. The only self-report is a single optional word about how speaking felt.
- It does not correct accents. Every relevant exercise tells the user to use their own accent. The targets are clarity and pacing within the speaker's own variety, not a shift towards any prestige variety.
The evidence position, step by step
We have tried to be accurate about which parts rest on established technique and which do not.
| Step | Framing in the app | Our position |
|---|---|---|
| Relax your face (external massage) | Presented on screen as relaxation and a transition into the session, explicitly labelled as not an articulation exercise. | This is a non-speech oral motor activity. The evidence-based systematic review found insufficient evidence that non-speech oral motor exercises improve speech, and we do not claim it does. It is included as a settling ritual, and the on-screen text says so. |
| Lip trills | Framed as a voice warm-up, not an articulation exercise. | Lip trills are a semi-occluded vocal tract exercise, and SOVT exercises do have evidence as a vocal warm-up. That evidence concerns voice, not articulation, and the app does not extend it. |
| Breath setup, jaw, lips, tongue tip, tongue back | Coordination practice, always with speech material attached. | Every one of these steps attaches real syllables, words or phrases rather than isolated non-speech movement, on the task-specificity principle. This was a deliberate design choice. |
| Clear sounds, word endings, sound combinations, clear sentences | Deliberate, over-articulated speech, then repeated naturally. | Deliberate or clear speech is a technique in wide clinical use. Its evidence base concerns intelligibility in people with diagnosed speech disorders, and does not transfer automatically to unimpaired speakers or to this particular sequence. |
| Easy pauses, word emphasis, guided reading | Rate control, phrasing and prosodic emphasis on supplied scripts. | Rate reduction and pausing are established techniques. Again, the evidence concerns clinical populations under clinical guidance, not this app. |
| Everyday speaking, fun finish | Transfer into unscripted speech and varied delivery. | Carryover to real speaking situations is the part of home practice most often skipped, so it is placed at the end of every routine rather than treated as optional. |
Safety
The safety page lists contraindications per step and the red flags that should send someone to a doctor rather than to practice: sudden slurred speech or unilateral weakness with explicit stroke messaging and emergency numbers, hoarseness beyond three weeks, pain on speaking or swallowing, progressive deterioration, and concerns about a child's speech development. Every exercise also carries a written caution, and the app repeatedly instructs users to stop if anything hurts.
Where it might be useful to you
- For someone on a waiting list who wants to do something in the meantime, with a clear statement that it is not a substitute for the assessment they are waiting for.
- As a warm-up before your session, so clinical time is not spent on it.
- For occupational voice users — teachers, lecturers, call handlers, clergy, actors — who want a habit rather than a referral.
- For people who will not or cannot access services: no cost, no account, no referral, works offline, and available in a growing number of languages.
Where it is probably not useful: anyone who needs individual assessment, anyone whose presentation is changing, anyone for whom a specific technique is contraindicated, and any child whose speech development is a concern.
A handout you can give someone
A printable one-page sheet, plain language, with the link, what it is, what it is not, and when to stop. Print it from the button below.
Help us make it better
We would like a registered speech and language therapist to review the routine for safety and framing, and we will name them and state exactly what they did and did not assess. We are not looking for an endorsement and will not describe any review as validation.
We would also welcome: corrections to anything on this page, native-speaker review of a language pack, and a recorded reference voice in any language. Open an issue on GitHub, or message Vyshak on LinkedIn.
The whole thing is open source at github.com/vyshakiyengar/daily_speech_exercise. Every exercise, instruction and caution is in plain text there, and you are welcome to open an issue.
← Back to practice