CW Training
CW Training is built for your ears, not your eyes. It sends a group of one to five random Morse characters as a single run, gives you a Farnsworth recognition pause, speaks every answer in the group in order, and either repeats that complete pair or begins the next group. The loop continues until you press Stop.
Group size, speed, and repetition are ranges rather than fixed settings, so the pause, the sending speed, how many characters arrive together, and how often they repeat all change from one group to the next.
During playback, the current character, answer, dots, and dashes never appear on screen. The character picker is visible only while configuring the session. No microphone permission is needed.
Start a Session
- Open CW Training in the Tools section of the sidebar.
- Set the Character speed range for the speed inside each character.
- Set the Spacing speed range for the longer recognition pause.
- Set the CW tone pitch and volume.
- Choose an answer voice.
- Choose letters, digits, punctuation, or any combination. Expand Choose individual characters to fine-tune those selected sets.
- Under Prompt shape, set Characters per group and the Extra repeats range.
- Press Play, listen, and name each character of the group in your head before the answers are spoken.
The defaults are an 18-20 WPM character-speed range with 8-10 WPM spacing, one character per group, and no extra repeats — so out of the box it is the same single-character drill it has always been, with only the speeds varying.
The CW pitch is adjustable from 400 to 900 Hz and defaults to 650 Hz. CW volume is adjustable from 20% to 100% and defaults to 60%. Both are ordinary single-value sliders. Stop playback before changing the controls; each session keeps the character set, ranges, tone, and voice it started with, and keeps drawing fresh values inside those ranges.
Ranges, Not Settings
Characters per group, character speed, spacing speed, and extra repeats are each a range with a low and a high thumb. Hamtrax draws a fresh value from every range for every group, and each value in a range is equally likely.
That variation is the point. A fixed speed teaches your ear the session's rhythm; a moving one teaches it the sound. Real operators never send at one exact speed, so practicing against a shifting target is both harder and closer to what you will hear on the air.
The two thumbs ride opposite sides of the rail: the high thumb above it, the low thumb below, each pointing at its stop. A thumb answers to its whole side of the slider — press or drag anywhere above the rail to move the high thumb, anywhere below it for the low — so there is no tiny handle to hunt for, and when a range collapses the two meet nose-to-nose across the rail instead of covering each other. Dragging one thumb into the other pushes the other along, so reshaping a range is one motion rather than two.
Every range may also collapse to a single value: put both thumbs on the same stop and the readout shows one number instead of a range. Holding a group length, a speed, or a repeat count steady is a deliberate drill — keep the thumbs together for that, close for gentle variation, or far apart for a workout. The one rule is order: the high thumb can never sit below the low one.
Character-speed stops are 12, 15, 18, 20, 25, 30, 35, and 40 WPM. Spacing stops are 5, 8, 10, 12, 15, 18, 20, 25, 30, 35, and 40 WPM, capped at the slowest character speed your range can draw — so lowering the character-speed low thumb also pulls the spacing range down if it no longer fits. Group-size stops are 1, 2, 3, 4, and 5.
Every slider is keyboard-operable: focus a thumb and use the arrow keys to move one stop at a time, or Home and End to jump as far as it can go — pushing the other thumb along if it stands in the way.
Practice Sets
- Letters: 26
- Digits: 10
- Punctuation: 18 (
. , ? ' ! / ( ) & : ; = + - _ " $ @)
The three circular controls are independent multi-select choices, so you can practice any combination. Expand Choose individual characters to include or remove specific characters inside every selected set. A partly selected parent shows a mixed state, and Hamtrax always keeps at least one character enabled.
The shuffle bag draws every enabled character once before reshuffling, and it prevents the last character of one cycle from immediately returning at the start of the next. Each character of a group is its own draw from that same bag, so full coverage carries across group boundaries rather than restarting per group.
Prompt Shape
Two controls decide what one prompt actually is.
Characters per Group
Set Characters per group anywhere from 1 to 5. Hamtrax sends that many characters as one continuous Morse run — separated by the normal Farnsworth character gap, exactly as an operator would key them — then pauses, then speaks every answer in order.
Copying a run is the real on-air skill: you have to hold each sound in your head while the next one is already arriving. A group of 1 drills pure recognition; a group of 4 or 5 drills the short-term memory that turns recognition into copying.
Leave both thumbs on 1 for classic single-character practice. Open the range — say 1-3 — and the group length itself becomes unpredictable, so you cannot settle into counting.
Extra Repeats
Set an Extra repeats range anywhere from 0 to 4. The number is how many extra times Hamtrax replays the complete group — the Morse run and all of its spoken answers — before moving to another group.
Put both thumbs on the same number for a fixed count: 0 sends each group once, 2 always gives you three passes at it. Open the range and the count is drawn fresh for each group instead — a 0-2 range means some groups play once and others play up to three times, so you can never count on a second listen, exactly like a live signal.
Every repeat of a group reuses that group's drawn speeds, so a repeat is genuinely a second listen at the same signal rather than a different one.
Answer Voices and Playback
CW Training includes three self-hosted answer voices:
- Allison · US — human-recorded American English and the default
- Jay · UK — human-recorded British English
- Original · US — the original synthesized Piper voice
The human voices use professionally recorded Asterisk prompt sets, trimmed and combined into compact Hamtrax audio sprites. The web app downloads and caches only the voice you select, so changing voices can require a connection the first time. The installed native app bundles all three packs.
On iPhone and iPad, Hamtrax coordinates its audio tools as one session. A running practice session uses media playback, so with no headphones or Bluetooth audio route selected, the tones and spoken answers play through the built-in speaker even in Silent mode. Hamtrax restores the phone's previous audio behavior after the last audio tool stops.
Minimizing the panel or clicking behind it does not interrupt practice. Open CW Training from the sidebar to restore the same session. Stop ends the session; closing the panel or signing out also stops playback and releases its audio resources.
Related Pages
- CW Decoder — reference for the microphone Morse decoder, currently unavailable in the app
- Band Plan — find CW segments on each band