Text to Video for briefs and scripts that need a complete visual story
Bring the approved words and the context behind them: audience, product, evidence, and visual direction. Turn that source into a complete branded project you can revise and adapt across the campaign.
This 31-second format example used a written brief containing a public GitHub Blog article URL and selected takeaways. Nia and Tevita are fictional AI hosts, and the generated workflow graphics illustrate the source.
How it works
Four steps from raw text to a video you'd actually post.
No template picker, no stock-clip keyword bingo, no slide-builder pretending to be a video editor. The text you already wrote becomes the script, and the script becomes a storyboard you can argue with before anything renders.
01
Paste the text or drop a URL
Script, prompt, release notes, product copy, blog draft, internal update. Up to 10,000 words. If your source is on a public URL instead, paste the link and ngram fetches the body with Firecrawl.
02
The agent rewrites it as a video script
Choose a direct narration or a scripted discussion from your approved text. For a two-host video podcast, name the hosts’ roles and the facts each turn should explain; the agent proposes dialogue and scene pacing for review before production.
03
Review the plan containing the script before render
Every scene shows the line of script, the visual direction, and the duration. Drop a section, rewrite a hook, swap a visual, or ask for a CFO cut in plain language, and every change ripples back into the script.
04
Review and export the video
Review the text video for its intended channel, including the crop and caption placement, then export one MP4 from Studio. Prepare and review another aspect ratio as a separate edit.
Output controls
Smart defaults from the text. Real knobs when you need them.
Script-first review
Read the full script before any visual is generated. Cut a paragraph, change a hook, or rewrite the CTA in plain English. Every edit re-flows the scene plan downstream.
Scene planner that reads structure
H1s, H2s, bullet lists, and even one-paragraph dumps all map to a hook + body + CTA storyboard. The agent re-paces longer text into video-friendly chunks instead of one slide per sentence.
AI Visuals per scene
Each scene gets a brand-matched image or short generative clip tied to that paragraph's meaning. No keyword-matched stock footage dragging the video into generic territory.
Brand Kit on every frame
Logo, fonts, colors and motion style pulled from your saved Brand Kit. Choose saved intro or outro slides for each video that needs them. The 50th text-to-video looks as on-brand as the first.
Voiceover from any saved voice
Read in a default ngram voice, your cloned founder voice, or a multilingual ElevenLabs voice. Pace and tone follow the script you approved.
Captions in your brand font
Turn captions on before export, with styling from the Brand Kit's caption preset. Edit a word in the script and that scene re-renders with the new caption. Captions start off; select your Brand Kit during creation to use its styling.
Prepare each channel edit separately
Set the intended channel and aspect ratio for the text video, then review the crop, caption placement, and CTA before exporting one MP4. Prepare another channel version as a separate edit.
Data handling for your team
Your pasted text, fetched URLs, and uploaded drafts live in your workspace, and you can delete your account and trigger a full data purge from Settings. Talk to sales about security, access controls, and data handling for your team.
The rest of ngram
What ngram does to a chunk of text that template builders don't.
Source is a URL, a doc, or a PDF? Pick the converter that matches the input.
Text to video is one node on ngram's script and storyboard pipeline. Every text-source converter on this list shares the same scene planner, Brand Kit, and review step. Prepare each channel version as a separate edit.
Trigger text-to-video where the writing already happens.
Wire the converter into your CMS, your CRM, your agent stack, or your publishing tools. Every integration ships with a working text-to-video template you can fork.
REST APIMCP serverWebhook subscriptionsFor API automation, send supported text or URL inputs with a self-service API key, retrieve the video job status, and subscribe to video.completed or video.failed.
How it compares
If you've been using something else for text to video.
Synthesia centers the avatar; the script lives in a side panel and the brand controls are template-level. Pictory and Lumen5 match each paragraph to a stock-clip library, scene-by-scene. ngram reads the structure, plans the storyboard you can argue with before render, and applies the Brand Kit per scene, so the output reads like your writing, not like a template.
Feature
ngram
Synthesia
Pictory
Lumen5
How the text is read
Structure + meaning. The agent rewrites text into a hook-body-CTA script.
Treated as avatar speech script
Sentence-by-sentence over stock clips
Sentence-by-sentence over stock clips
Storyboard review before render
Full scene-by-scene plan, editable in plain language
Scene list, limited script editing
Scene cards, limited script edits
Inline timeline, no script-level review
Visual generation
AI Visuals matched to scene meaning per Brand Kit style
Avatar over template backgrounds
Stock-library matching
Stock-library matching
Brand application
Brand Kit for logo, fonts, colors and motion. Choose a compatible outro separately.
Template-based, limited per-scene control
Brand presets, limited per-scene control
Template-based, limited per-scene control
Channel-specific editing workflow
One reviewed MP4 per channel-specific edit
One ratio per render
One ratio per render
One ratio per render
Voiceover
ElevenLabs voices + cloned founder voice, any supported language
Avatar voice library
Limited TTS voices
Limited TTS voices
Audience-focused script revisions
Ask for more concise, technical, or direct copy, then review and render the revised edit separately
Manual rework
Manual rework
Manual rework
Source content control
Source text stays in your workspace, with account-level data purge from Settings
Indefinite retention
Indefinite retention
Indefinite retention
API + agentic access
REST API with self-service API keys; MCP, Zapier, and n8n for documented workflows
Paste a script, prompt, or draft up to 10,000 words. The agent reads the structure, rewrites it as a hook-body-CTA video script, and builds a scene-by-scene storyboard. You review and edit the storyboard in plain language. The whole flow runs in under five minutes from paste to publish. After review, Studio exports the text video as one MP4. Prepare and review another channel version as a separate edit.
What kinds of text work best for text to video conversion?
Anything with structure converts cleanest: launch scripts, product copy, blog drafts, release notes, marketing briefs, internal updates, ad hooks. Plain paragraphs work too, and the agent picks up sentence-level topic shifts and chunks them into scenes. Longer drafts are compressed automatically and surfaced for review in the storyboard.
Can I use a PDF instead of pasting the text?
Yes. Choose Upload PDF and select the document containing your brief, script or source material. Review the proposed video plan in ngram before starting production.
Does text to video use stock footage or AI-generated visuals?
AI-generated visuals matched to the meaning of each scene, styled against your Brand Kit. Stock-library matching is what Lumen5 and Pictory do; ngram avoids it because keyword-matched stock makes every text-to-video look the same. If you want stock or your own footage in a scene, you can swap it in during storyboard review.
How long is the output text to video?
Most outputs land between 30 seconds and 3 minutes, depending on how much text you paste in. You can ask the agent for a shorter cut ("compress to 60 seconds") or a longer cut ("hold every section, expand the demo") and the storyboard re-paces around the new target.
What input languages and output voices are supported?
Source text in any of the major languages handled by frontier LLMs. Voiceover is generated via ElevenLabs and MiniMax across a broad voice library, including cloned voices uploaded to your team's Custom Voices library. Captions and on-screen text follow the script's source language by default; you can re-render in another language without redoing the storyboard.
Can I prepare another audience or channel edit from the same source text?
Yes. Keep the text as source material and ask for a shorter, more direct, or more technical revision in chat. Review the revised script, pacing, crop, and captions, then render that channel edit separately.
What aspect ratios and file formats can I export?
Studio exports the reviewed text video as one MP4. Set the aspect ratio for that edit, then check its crop and caption placement before export. Prepare and review a separate edit for another channel. If a destination requires MOV, WebM, or MKV, use Video Converter as a separate one-file, one-output step; use Video to GIF for a GIF. Output is 720p on Basic, 1080p on Premium, and 4K on Ultimate. Share a public watch link on Free.
What happens to my pasted text after conversion?
Your pasted text, fetched URLs, and uploaded drafts live in your workspace, and you can delete your account and trigger a full data purge from Settings at any time. Talk to sales about security, access controls, and data handling for your team.
Can I integrate text to video into our existing tooling?
Yes. Zapier and n8n connectors handle no-code triggers from CMS, CRM, and project tools. There's an MCP server for documented agentic tools and a REST API with a self-service API key; public webhook subscriptions emit video.completed or video.failed. Pricing is credit-based: The base rate for premium AI video generation is 10 credits per second; a cheaper model costs 3 credits per second from your plan allowance. API-key video creation requires Basic or higher and credits.
How is ngram different from other text to video generators?
Most text to video tools generate first and let you edit after. ngram plans first: you see and approve the plan containing the script in plain language before business video renders. The output also applies your Brand Kit to every scene, so a 50th text-to-video looks as on-brand as the first. Template tools don't do either.
Can text to video create a two-host video podcast?
Yes. Paste the approved article text, outline or discussion brief and state the evidence the hosts may use. In Prompt to video, describe two fictional on-screen hosts and the source facts in Video brief. Select Conversation under Start from a style with Opus 5.5, Presenter Auto and Voice Auto, then review both voices and speaking turns in the plan before production. Give the questioner a useful question that prompts a supported explanation. Keep invented hosts separate from real authors, and check every factual response against the supplied material. Adapt the roles in the two-host template to your approved text. For a worked source-and-role outline, follow the article-to-podcast tutorial.
Text → Video
Ready to turn the next paste into a branded video?
Drop in the text. Review the plan containing the script, then generate your video. Prepare and review each channel edit separately before exporting one MP4.
Give Claude the buyer question and approved product evidence. Review ngram’s answer and visuals, then create a focused video with one practical next step.
One buyer question and current approved product evidenceA concise answer video grounded in the approved source
Ways to run: Claude
Each new lesson resource or classroom introduction
Give Claude the learning goal, key vocabulary and a question to leave open. Review ngram’s scene plan, then create a short preview that invites viewers into the lesson.
A lesson objective, vocabulary, preview example and resource titleA short educational teaser with an open question and resource title
Give Claude your product facts and real photos. Review ngram’s plan, then create a short listing video that keeps the product’s appearance and attributes accurate.
Current product facts and approved photosA source-faithful product-page video
Give Claude the property brief and photos you want to show. Review ngram’s narration and scene plan, then turn the selected features into a video for your listing campaign.
A property brief, selected photos and a clear viewing actionA property video built around the supplied details and visuals
Give Claude a verified public account update, relevant product proof and one proposed next step. Review ngram’s plan, then create a short introduction you can check before sending to the prospect.
Verified public account news, approved product proof and one next stepA concise prospect-specific introduction ready for review
Give Claude the questions, approved answers and speaking roles. Review ngram’s Conversation plan, then create an exchange with distinct presenters and voices.
Approved questions and answers, speaker roles and supporting product factsA two-speaker product explanation in Conversation style
Ways to run: Claude
Each approved campaign, recipient segment or newsletter edition
Give Claude the campaign brief, recipient audience and next action. Review ngram’s video, then pair a public thumbnail with a ready viewing link for your email.
Approved campaign copy, audience details, imagery and a destinationA campaign video and linked thumbnail ready for an email preview
Ways to run: Claude
Each onboarding cycle or security-awareness refresher
Give Claude the script your team has reviewed and the concept you want to illustrate. Check ngram’s scene plan, then create a training video that keeps the wording intact.
A reviewed training script and visual analogyA narrated security lesson with the exact supplied script
Ways to run: Claude
Each recurring onboarding question or approved process update
Give Claude the approved customer instructions for one onboarding step. Review ngram’s explanation, then create a video that tells customers what to prepare and what to do next.
Approved onboarding instructions and demonstration assetsA focused customer instruction video
Ways to run: Claude
Each audience, positioning update or new product explanation
Give Claude the audience, approved positioning and supported product facts. Review ngram’s plan, then create a concise explanation of how the product’s main concepts fit together.
An approved positioning brief and current product sourcesA source-backed software explanation with clear labeled visuals
Give Claude the policy memo, including its effective date and required employee action. Review ngram’s plan and share the finished video alongside the written policy.
An approved policy change and effective dateA narrated employee explainer
Give Claude your roadmap and the feedback you need from stakeholders. Review ngram’s plan to keep available features, work in progress and tentative ideas distinct, then create the update.
Current priorities, availability and dependenciesA narrated roadmap briefing
Give Claude a published article and the ideas your campaign should cover. Review a short video plan, create it with ngram and lead viewers back to the full post.
A published article and selected takeawaysA short video linked to the full article
Ways to run: Claude
Each brand campaign, audience or refreshed visual brief
Give Claude the brand point of view, approved narration and permitted imagery. Review ngram’s plan, then create a short story that keeps the visual treatment faithful to those sources.
An approved brand brief and permitted source imageryA concise brand story with faithful source visuals
Ways to run: Claude
Each client campaign, approved message or creative variant
Give Claude one client’s approved message, visual assets and campaign placement. Review ngram’s scene plan, then create an ad your team can check and send for client review.
Client facts, permitted brand assets, approved narration and one placement briefA client-specific video ad ready for review and export
Give Claude a dated comparison and the buyer’s actual requirements. Review a fair, source-backed video plan, then create an explanation of the next step with ngram.
A dated battlecard and buyer requirementsA focused comparison video with cited sources
Give Claude one module’s objective, worked example and practice task. Review ngram’s teaching sequence, then create a lesson learners can follow and apply.
One module outline, supporting material and a worked exampleA complete lesson with an actionable practice task
Give Claude your help article and the question customers keep asking. Review ngram’s plan, check the troubleshooting steps and add the finished video to your support answer.
A documented symptom, checks and resolutionA reusable troubleshooting video
Give Claude the approved scope and terms. Create a narrated walkthrough with ngram, check the plan against the document and share the finished explanation with your buyer.
An approved proposal and its commercial termsA narrated buyer walkthrough
Ways to run: Claude
Each approved organizational announcement or internal initiative
Give Claude the approved news, intended staff audience and employee action. Review ngram’s plan, then create a concise announcement people can understand and act on.
Approved company news, audience details and employee actionsAn internal announcement video with a clear next step
Ways to run: Claude
Each recurring patient question or approved guide update
Give Claude the patient guide and the lesson you want to explain. Review ngram’s narration and illustrations, then create a short video patients can watch alongside the full guidance.
Approved patient information and a defined educational lessonA short, source-based patient education video
Give Claude the buyer’s priorities, demonstrated screens and open questions. Review ngram’s plan, then create a concise recap that keeps the product facts and proposed next step clear.
Reviewed demo notes, genuine demonstrated screens and outstanding questionsA buyer-specific recap with a clear evaluation step
Give Claude the findings, source attribution and decision your team is considering. Review ngram’s plan, then create a visual briefing that connects the evidence with the next research task.
Reviewed findings, attributed quotes and a proposed next stepA narrated stakeholder briefing with source credits and a clear next test
Give Claude the services, audience and booking action from your page. Review ngram’s plan, then create a short promotion around the details customers need to choose their next step.
Service-page copy, approved service details and a booking actionA short service promotion for a social feed or service page