Ask your videos
anything.

Ask what was said or what's shown on screen. Upload a video, paste a YouTube link or drop in audio, and get transcripts, summaries and scene breakdowns included.

Start for freeSee how it works

200 free tokens every month. No credit card required.

Works with

  • MP4, MOV, MP3, WAV
  • YouTube links
  • ChatGPT app
  • MCP server
  • REST API
chat.videoinsights.ai/media-studio
VideoInsights Media Studio showing an automatic scene index, entities and relations for an uploaded video
Chat · onboarding-call.mp442:17

Ask anything about the video…

How it works

From raw footage to answers in three steps

No setup, no editing timeline. Upload, wait a moment, then start asking.

Upload your first file
  1. 01

    Bring your media

    Drag in a video or audio file, or paste a YouTube link. Files are encrypted on upload.

    MP4, MOV, MP3, WAV, YouTube

  2. 02

    We analyze it

    Transcription runs immediately. Add visual analysis when you want to know what happens on screen, not just what is said.

    1 token per minute of audio

  3. 03

    Ask, export, build

    Chat with the content, export subtitles and summaries, or pull everything through the API and MCP server.

    Web app, API, MCP, ChatGPT

Platform

Everything your media knows, one question away

Chat is the fastest way in. Export and visual tools are there when you need more than an answer.

Chat with your videos

Ask a question and get an answer with timestamps you can click. Want to know when the speaker mentions pricing? Just ask.

Chat · onboarding-call.mp442:17

Ask anything about the video…

Sees what happens on screen

Beyond the transcript: scenes, people, objects, gestures and on-screen text, indexed by time.

  1. 0:00

    Man in black carrying box

    walkingcarrying box
  2. 0:08

    Green outfit, adjusting

    holding boxlooking down
  3. 0:20

    Examining needle

    glovesmedical tray
  4. 0:24

    Checking laser door

    standingsign

Upload your own files

MP4, MOV, MP3, WAV and more. Up to four hours per file on Professional.

YouTube, no download

Paste a link. Transcript, metadata and comments are pulled for you.

Subtitle export

SRT and VTT with speaker labels, translated into the languages you need.

AI summaries

A short recap, a detailed outline with timestamps, or bullet points. Your pick.

SEO metadata

Titles, descriptions, tags and schema.org markup generated from the content.

Privacy first

Encrypted at rest, auto-deleted after your retention period, never used for training.

For developers

API, Python SDK and an MCP server

Every paid plan includes REST API access. Point Claude, Cursor or ChatGPT at the MCP server and your media library becomes a tool.

API docsPython SDK
# pip install videoinsights
from videoinsights import Client

vi = Client(api_key="vi_live_…")
media = vi.upload("all-hands.mp4")

answer = vi.ask(media.id,
    "When is the pricing change mentioned?")
print(answer.text, answer.timestamps)
Who it's for

Too much recorded media. Not enough time.

Different industries, the same problem. Here is how teams use VideoInsights day to day.

Explore use cases

Law enforcement

Search hours of bodycam footage by event.

Media production

Find the usable moments in raw footage.

Education

Make lecture recordings searchable.

Legal

Transcribe depositions, cite by timestamp.

Teams

Turn meetings into action items.

Moderation

Flag content for human review at scale.

New: VI Edge

For data that cannot leave the building

A rack-mounted appliance with the full VideoInsights stack and an offline LLM. No cloud, no subscription, no egress. Now taking waitlist signups.

Join the waitlist
VI Edge rack-mounted appliance
200 free tokens every month

Try it with your next video

Sign up, upload one file, and ask it something. If it takes longer than a minute to see the value, tell us.

Start for freeSee pricing