AudioChunker

Audiochunker — slice any audio into perfectly timed clips, instantly

Audiochunker is a lightweight Pinokio app that takes any audio file and chops the whole thing into fixed-length clips with one click. Drag in a track, pick a length on the seconds slider (10 / 20 / 30 / 40 / 50 / 60s), and it splits the entire file into back-to-back segments — then export them all at once as a single .zip, or grab any clip on its own.

What it's built on

ffmpeg under the hood (segment muxer with stream copy -c copy) — so clips are cut without re-encoding and keep the source's exact codec and quality. No quality loss, no waiting on transcodes.
Streamlit UI — clean web app with per-clip preview players, timestamps, individual downloads, and a one-click "Export ALL (.zip)".
CUDA-ready torch baked into the installer for GPU-accelerated workflows.
Packaged the Pinokio way (install.js / start.js / torch.js) — one-click install and launch.
What it does well

Separates audio into clips fast — a full track is chunked in seconds.
Any format in (mp3, wav, flac, m4a, ogg, aac, opus, wma…), original quality out.
Consistent, predictable cut points — great for batching.
Why it's powerful in a pipeline Pair Audiochunker with a video generator like LTX: feed it evenly-cut audio segments and generate a fresh visual for each clip, then stitch them back in order. It's a fast way to switch scenes in a music video on the beat — chunk the song into 10–60s sections, generate or swap a new LTX scene per section, and you get a multi-scene music video that changes in time with the audio. Same trick works for reels, lyric videos, podcast highlight cuts, or any "one clip → one generated shot" workflow.