Media Transcriber - Audio, Video & Podcast to Text, SRT, VTT
Pricing
from $2.00 / 1,000 audio minute (your api key)s
Transcribe audio, video and podcast feeds to text with timestamps, SRT and VTT. Bring your Groq/OpenAI key for full-length Whisper, or use the built-in small model. Language detection, English translation. From US$2 per 1,000 audio minutes.
id
Optional
Stable id of the media item (hash of the episode guid or URL)
sourceUrl
Optional
URL given in the input (a media file, share link or podcast feed)
mediaUrl
Optional
Media URL that was downloaded (query string removed)
title
Optional
Episode / file title
feedTitle
Optional
Podcast feed title when the source was an RSS feed
publishedAt
Optional
Episode publication date (ISO 8601) when known
status
Optional
ok or failed
error
Optional
Reason when status is failed (failed items are not charged)
engine
Optional
groq, openai or local
model
Optional
Model used, e.g. whisper-large-v3-turbo
task
Optional
transcribe or translate (to English)
language
Optional
Spoken language, ISO 639-1 code (null when translating)
languageProbability
Optional
Detection confidence 0-1 (local engine only)
durationSecs
Optional
Seconds of audio transcribed
billedMinutes
Optional
Audio minutes charged for this item
truncated
Optional
True when only the first part of the media was transcribed (local 10-minute cap or maxMinutesPerItem)
text
Optional
Full transcript as plain text
segments
Optional
Timestamped segments: id, start, end (seconds), text
words
Optional
Word-level timestamps (API engines, when wordTimestamps is on)
srt
Optional
Transcript as SubRip (.srt) subtitles
vtt
Optional
Transcript as WebVTT (.vtt) subtitles
txtUrl
Optional
Download link of the .txt file in the run key-value store
srtUrl
Optional
Download link of the .srt file
vttUrl
Optional
Download link of the .vtt file
processedAt
Optional
Processing time (ISO 8601)
