Podcast Transcript & Subtitle Generator

Show notes and blog posts: a transcript turns an episode into text that listeners can skim and search engines can read, which audio alone can't give you.

YouTube: uploading an SRT file with a video podcast gives captions with names spelled right, instead of YouTube's automatic ones.

Clips: cut a strong minute out of the episode with the Video Trimmer, then add animated captions to post it on Reels, Shorts or TikTok.

Accessibility: listeners who are deaf or hard of hearing can follow the whole conversation.

Accurate mode (the default): your browser reads the audio out of the episode and sends only that audio to our server, where Whisper large-v3-turbo transcribes it (at Groq with data retention turned off, or the same model at DeepInfra, which doesn't store audio either). Nothing is kept. A video file never leaves your device.

Accurate mode takes episodes up to 15 minutes. Longer episodes are transcribed on your device with a smaller Whisper model: free and private, but slower and less accurate (on a 2020 laptop, roughly 4–5 times faster than the audio plays).

The transcript doesn't say who is speaking. In the text download it's split into paragraphs at pauses; the SRT and VTT files keep every line with its timing.