Seply

Seply does one thing: speaker separation. Upload a podcast, interview, meeting or video recording, let it auto-detect the speakers (or set the count manually), and download each voice as its own time-aligned WAV track. There's a standard mode for orderly conversation and an advanced overlapping-speech mode for the interrupt-heavy conversations where simpler tools fall apart.

Per-track WAV output — not just a transcript — means editors can remix levels, cut crosstalk, or run each speaker through their own transcription pass with noticeably better accuracy.

Who it's for: podcast editors cleaning up multi-person shows, journalists and researchers processing interviews, and anyone whose transcription tool keeps merging two speakers into one paragraph. Usage-based pricing around $0.06-0.08 per minute of audio. The single-purpose focus is the strength; if you need full editing, pair it with your DAW rather than expecting it to be one.