Free tool
Split any audio into sentences
Drop in a file, get clean sentence segments with timestamps — transcribed by AI entirely in your browser.
Cutting audio into sentences by hand is miserable work: scrub, guess where the sentence ends, cut, repeat — a ten-minute file takes an hour in a waveform editor. ListenSlice does the whole job automatically. Whisper AI transcribes your file with word-level timestamps, then slices it into sentence segments you can play, loop, and export.
Unlike server-based transcription tools, everything runs locally in your browser via WebGPU. Your audio never leaves your device — which also means there is no upload wait, no file-size anxiety about private recordings, and no terms-of-service risk with copyrighted material you own.
live demo — real audio, real scoring
How it works
01Drop your audio in
MP3, WAV, or M4A up to 100MB. Or try the built-in sample first.
02AI slices it
Whisper transcribes locally and cuts the audio at sentence boundaries.
03Use the segments
Play sentence by sentence, practice dictation, or export SRT/TXT/JSON.
Millisecond-accurate segment playback
Each sentence gets precise start and end timestamps, and the built-in player uses the Web Audio API to play exactly that window — no bleeding into the next sentence, no cutting off the last word. Click any sentence in the list to hear just that sentence, as many times as you need.
Built for language learners, useful for everyone
ListenSlice was built for sentence-by-sentence listening practice, but the splitter is general-purpose: podcasters clipping quotes, editors pulling soundbites, researchers coding interview recordings. Pro users can export the segments as SRT subtitles, plain text, or JSON with timestamps.
Frequently asked questions
Does my audio get uploaded to a server?
No. Transcription runs inside your browser using WebGPU (with a WASM fallback). The audio file never leaves your device — there is no upload endpoint at all.
How accurate is the sentence splitting?
Segmentation uses Whisper word-level timestamps and punctuation, which is highly reliable for clear speech. Very long run-on sentences or heavy crosstalk can occasionally merge; you can navigate by segment and replay any span precisely.
What audio formats and lengths are supported?
MP3, WAV, M4A, AAC and OGG up to 100MB. The free tier processes files up to 15 minutes; Pro removes the duration limit.
Can I export the sentences with timestamps?
Yes — Pro accounts export SRT subtitles, plain-text transcripts, and JSON with per-sentence start/end times.
Ready to try it with your own audio?
free to start no sign-up audio never uploaded
related:Turn any audio into dictation exercises·A player that repeats one sentence at a time·Practice listening with your own MP3 files