SpeakerSplit - Video and Audio Tools

Audio toolkit that splits and edits podcasts, interviews, meetings, music and video. Connect it and ask in plain words. Claude starts the job, waits for it to finish and gives you the download links.

What it can do:

• Split speakers: one audio track per person, plus a transcript labeled with each speaker's name.
• Transcribe: a transcript with each speaker named, as plain text, Markdown, SRT or VTT subtitles, or JSON.
• Remove vocals: the voice alone and the music alone, for example for karaoke.
• Split into stems: vocals, bass, drums, piano, guitar and other as separate tracks.
• Clean up audio: remove background noise from speech.
• Remove silences: cut pauses and filler words such as "um" and "uh" out of audio or video.
• Add captions: burn the spoken words into a video, in time with the speech.

Good for podcast editors who need each voice on its own track, producers writing show notes and pulling quotes, and anyone who needs a clean record of who said what.

Each job uses credits from your SpeakerSplit account, based on the recording's length. Each tool's description gives its price, and credits for a failed job are returned. Checking a job, reading a transcript and checking your balance are free. Speaker splits and transcripts take recordings up to 4 hours, other tools up to 90 minutes (videos up to 30 minutes for captions and silence removal). Results can be downloaded for 30 days.