Audio Transcription (Whisper)

Speech to text for a list of audio or video file URLs, using Whisper (faster-whisper). You get the text, timed segments, and SRT or VTT subtitles for each file.

Run it on the Apify Store

What it does

• One result row per file with its own status: ok, failed (with the reason) or too_long. One bad link does not fail the run.
• Models base (quicker, less compute) and small (more accurate, slower); language auto-detected or set.
• Output: plain text, timed segments, SRT, VTT.
• A cap on total audio minutes per run, so a big list cannot surprise you.
• Callable by AI assistants through Apify's MCP server.

Good to know

Direct links to audio or video files only (mp3, wav, m4a, ogg, mp4 and anything ffmpeg reads, up to 500 MB each). YouTube, TikTok and Instagram page links are refused. No speaker labels. Accuracy depends on audio quality and language; check important transcripts.

Price

No tool fee yet: you pay only Apify's platform usage for your run. In our test, 4.4 minutes of audio on base cost under US$0.02 of usage.

A bug or a feature you need? Open an issue on the Store page.