AI voice coach · completely free
AI voice coach for speaking, not singing
This is daily training for the voice you use in meetings, calls, interviews and presentations. No scales, no songs, no tuning. You breathe, warm up, work on resonance and articulation, then speak — and the feedback afterwards is based on what was actually measured in your recording, not on flattery.
Speaking voice training, not singing lessons
Most voice material online is written for singers. It talks about range, pitch accuracy and performance. None of that is the problem when a colleague asks you to repeat yourself on a call, or when your voice goes thin and fast three minutes into a presentation.
SpeakFather trains the speaking voice directly: steady breath so sentences do not collapse at the end, a warm relaxed sound instead of a tight one, consonants that survive a poor connection, and a pace that gives the listener time to follow you. A few exercises — humming, lip trills, sirens — are borrowed from singers, because they are the gentlest way to wake a voice up. You are never asked to sing.
Don Carlos, the coach, explains each exercise out loud before you try it, and an animated demonstration shows the movement. You watch once, then follow at your own comfortable pace.
The daily workout
Pick five, ten, fifteen or twenty minutes. The plan adapts to the area you have trained least, so you do not spend three weeks on breathing and never touch articulation.
Breathing
Diaphragmatic breathing, 4-4 control, breath support for speaking.
Vocal warm-ups
Humming, lip trills, sirens and gentle vowel transitions.
Pitch & resonance
Pitch glides, resonance awareness and speaking-pitch exploration.
Articulation
Consonant drills, vowel clarity, tongue twisters, precision speaking.
Words & sentences
Five levels, from single words to professional delivery.
Speaking practice
Timed speaking challenges with measured feedback afterwards.
Inside each module
Breathing
Almost every speaking problem that sounds like nerves is really a breathing problem. When the breath is high and shallow, sentences start strong and die out, the pace speeds up, and the throat takes over work the lungs should be doing.
The breathing exercises are diaphragmatic breathing, a 4-4 control pattern, sustained exhale and breath support for speaking. A live animated guide shows where each hand goes — one on the chest, which stays still, one below the ribs, which moves — and a timed circle with optional breath sound leads you through each inhale and exhale.
Once a low breath is comfortable in silence, the coordination drills attach it to speech: one thought per breath, with a marked point to breathe, so you stop running out of air mid-sentence.
Vocal warm-ups
Speaking cold is why first calls of the day sound rough. The warm-ups — gentle humming, lip trills, tongue trills and easy vowel transitions — ease the voice into use without pushing it.
Each one has its own moving demonstration: lips closed with buzz rings at the lips and nose for humming, fluttering lips with a steady stream of air for trills, a morphing mouth shape for vowels. Soft is correct. If the throat is working hard, the instruction is always to do less, not more.
Pitch & resonance
Two people can say the same sentence and one carries across a room while the other does not. That is resonance: where the sound vibrates, rather than how loud it is.
Pitch glides, sirens and resonance awareness help you feel the difference between a thin voice pressed out of the throat and a full one that rings in the chest and face. A speaking-pitch exploration then helps you find the range that is comfortable to hold for a whole meeting.
Pitch is reported as an estimate from the audio, and only when it can be detected reliably. It is never dressed up as a precise measurement, and there is no target you are supposed to reach.
Articulation
Being asked to repeat yourself is rarely about accent. It is usually about consonant endings disappearing and words running together when you speed up.
Consonant drills, vowel clarity work, tongue twisters and precision speaking sharpen the edges of words. The rule is precise first, natural second: exaggerate while practising, then let it settle into normal speech, where it shows up as clarity rather than as effort.
Speaking practice
The last module puts it all together. Five levels take you from single words to phrases, sentences, paragraphs and full professional delivery, with a read-along pace guide and a marked breath point.
Then come timed speaking challenges: you answer a prompt out loud, and afterwards you see what was measured — pace, fillers, pauses, volume consistency — with one specific thing to fix next time. Levels run from Foundation to Executive Voice, and a streak keeps you honest about showing up.
What the voice coach measures
Everything below is calculated in code from your own audio and transcript, then handed to the coach as fact. Anything that cannot be measured honestly is simply not shown: no invented voice quality score, no accent rating, no clinical claims.
- Speaking rate in words per minute
- Filler words and filler rate
- Audible pauses, pause count and longest pause
- Volume consistency and volume drift across the recording
- Estimated pitch and pitch range, when it can be detected
- Time spoken and time practised over the week
Values that can only be approximated, such as pitch, are labelled as estimated. Posture is coached with plain guidance and explicitly not measured. Because the same numbers are recorded every session, progress is something you can check rather than something you have to feel, and you can compare an early attempt with a recent one side by side.
Voice training should feel comfortable. Stop if you experience pain, dizziness, significant throat discomfort or vocal strain, and speak with a qualified professional if symptoms persist.
Frequently asked questions
Is this singing training?
No. Every exercise here is built for the speaking voice: meetings, calls, presentations and interviews. Humming, lip trills and sirens are borrowed from singers because they warm the voice gently, but nothing asks you to sing, hold a tune or hit a note.
Do I need any equipment?
A browser and the microphone you already have. No headset, no interface and no quiet studio; a normal room is fine, and the exercises are designed to be done sitting or standing at your desk.
How long does a voice training session take?
Choose five, ten, fifteen or twenty minutes. Five minutes a day is enough to build the habit, and the workout adapts to the area you have trained least.
Will it tell me my voice is unhealthy?
No. SpeakFather is training, not a medical assessment, and it makes no clinical claims. Voice training should feel comfortable — stop if you experience pain, dizziness, throat discomfort or strain, and speak with a qualified professional if symptoms persist.
What happens to my voice recordings?
They are measured and then discarded. Recordings are only kept if you switch retention on yourself, so you can compare an early attempt with a later one.
Train for ten minutes a day
Free, with no paid plan. A coach who tells you what to fix next, and numbers that show whether it worked.
Start free