Story AI runs in three independent stages. You pick which ones you need — and you see exactly what it costs before you run anything.
Transcribe gives you word-level timestamps and speaker labels. Emotions gives you sentiment, story structure and chapter beats. Visual gives you scene descriptions, keyframes and B-roll suggestions.
Each service is priced per minute of footage. No subscriptions, no minimum spend.
How it works: Select the services you need, drag the duration slider to match your footage length, and the total updates in real time. Presets like Newsroom and Full Stack show common combinations used in production workflows.
What each stage does
Stage 1 — Transcribe
Whisper-based speech-to-text with word-level timestamps. Optional speaker diarization (who said what), translation to any language, and auto-subtitles. Free tier: basic transcription is included at no cost.
Stage 2 — Emotions
Sentiment analysis, story summary, narrative outline, key quotes, chapter detection and keyword extraction — all from the transcript. This is what turns raw speech into searchable story metadata.
Stage 3 — Visual
Scene detection, keyframe extraction and GPT-4V visual descriptions for every segment. Optional B-roll keyword suggestions (integrated with Storyblocks and Pexels) and AI music scoring.
You only pay for the stages you actually run. A basic transcription job on 10 minutes of footage costs less than a coffee. A full Visual + Emotions pass on a 60-minute documentary costs a few euros.
Sign in to see your credit balance reflected live in the calculator.