Upload a track or listen on another device, tap syllables, then generate lyrics from this page.
Drag & drop an audio file here
MP3 or WAV
Your audio stays in this browser tab’s memory only. It is never uploaded, never sent to a server, and never stored. The file never leaves your computer.
Listen on another device. Play starts a timer here so you can tap syllables as usual. Replay flashes those taps with no audio.
On a phone, tap the big button for each syllable while Play is running. New line starts the next phrase.
| Phrase | Syllables | Visual phrase | Start | End | Duration |
|---|
When enabled, Enter records a syllable and starts the next lyric line with that tap. When disabled, Enter only starts the next line.
When enabled, checks that generated lyrics match the rhythm pattern. Extra or missing syllables are marked in red. Off by default.
When “Say syllables out loud” is on, the browser voice starts this many milliseconds before each tap, to cover TTS delay. Default 50ms. Uses the device’s built-in voices. English voices cannot speak Malayalam or other non-Latin scripts — install that language’s speech voice in Windows if Test voice is silent.
Volume of the uploaded track in this tab. Turn this down if the spoken syllables are hard to hear over the music.
Volume of the spoken lyrics. That voice is separate from the waveform and may use a different Windows output device. Test voice speaks a sample in the lyric language (or Target Language). If Malayalam stays silent, install a Malayalam speech voice in Windows Settings → Time & language → Speech.
When enabled, subtracts this delay from each tap so markers line up with what you heard. Use ~500ms for many Bluetooth headphones. Off by default.
Extra offset applied when playback is running, for human reaction time and app processing. Default 100ms.
| Spacebar or Play | Play / Pause the track or the external-audio timer |
| Shift + Space | Replay taps from the first recorded syllable |
| Ctrl + Space | Play from the beginning of the waveform |
| Any other key or Tap syllable | Record an unmarked syllable (while playing) |
| ; or : | Long syllable (oval) while playing, or retag the selected tap |
| ' or " | Short syllable (diamond) while playing, or retag the selected tap |
| 1 / 2 | Set the selected syllable to short (diamond) / long (oval) |
| Click a tap marker | Choose unmarked, short (diamond), long (oval), or freeze that syllable |
| Click a lyric box | Type or edit the syllable for that tap. Empty boxes are placeholders until you add lyrics |
| Ctrl-click or right-click a syllable | Freeze / unfreeze that syllable (green outline). Frozen syllables stay put when you re-generate lyrics |
| ← / → | Seek backward / forward 2 seconds |
| , / . | Select the previous / next syllable tap on the waveform and phrase diagram |
| Enter or New line | Phrase boundary (X in the rhythm pattern). Optionally also taps, from Settings |
| Backspace | Delete last tap · Pause · Seek to the previous tap on the waveform |
| Start | Jump to the beginning of the waveform |
| First tap | Jump to the first recorded syllable on the waveform |
| Say syllables out loud | When enabled, playback speaks each lyric at its tap (starts early by the Settings head start). Volume is in Settings; Test voice checks the computer’s speech output |
| Clear & Start Over | After Yes, clear lyrics and taps and return to the start of the waveform |
| Export MP3 with metadata | Save a copy of the loaded MP3 with tap data in an ID3v2 tag (MP3 only). Loading that file restores the taps. |
| Right-click / long-press a waveform tap | Delete that syllable on the waveform (asks Yes / No) |
| Wheel on waveform | Zoom in / out · Shift+wheel scrolls sideways |