audioJS

Speech prosody editor

Shape how a sentence sounds. Adjust its pitch, emphasis and pace.

Drop a speech recording here
or click to choose, or paste with Ctrl+V
Experimental · one voice · up to 60 seconds · mono output

Choose a short, clean recording to begin.

Nothing is uploaded. Everything runs in this tab with @audio/pitch-yin, @audio/shift-formant, @audio/stretch-wsola, @audio/encode-wav.

Edit delivery, one phrase at a time

Select a word or phrase, reduce excessive pitch variation, shift its emphasis or adjust its duration. Changes are rendered from the original recording, so undoing an edit does not require reversing an audio effect.

This first version uses frame-level YIN detection, a formant-compensated pitch shifter and WSOLA time stretching. Timing edits stretch the whole selection, including consonants; automatic phoneme protection and linguistic intonation correction are future work. WAV export contains audio only, without source tags.

Questions

Can this automatically fix the tone of a question?

The rising and falling controls apply a pitch ramp to your selection. They do not infer sentence meaning. Listen and adjust the selection and pitch to match your intended delivery.

Will it preserve the voice?

Pitch shifting compensates for formant movement, but edits can still sound processed. Breathy or creaky speech, inaccurate pitch detection and large changes are harder. Compare with the original before saving.

Why are parts of the curve missing?

Silence and unvoiced consonants do not have a reliable fundamental pitch. Those regions stay unpitched. Very short clips may also have too little audio for detection.

More audio utilities