JoggAI
AICreate talking-avatar and template videos, lip-sync clips, and ad scripts with JoggAI, dubbing into more than 40 languages. Agents produce localized video variants without a shoot per market.
What This Integration Enables
JoggAI is a video production line, and the interesting part is the end of it. Create Avatar Video renders a talking presenter from a written script that JoggAI voices, or from audio you already have, with the avatar filling the frame, sitting beside your content or floating picture in picture. Create Template Video fills the declared variables of one of your templates, which is the fastest route to something on brand without designing anything. Generate AI Script writes marketing scripts from product information at a 15, 30 or 60 second runtime in a chosen style, from a reassuring opener through storytelling, a discovery angle, a data-led pitch or a top three reasons rundown, and returns several variants so a person can pick rather than accept.
Then there is Translate Video, which dubs a finished video into another language with an AI voiceover, optional translated subtitles, and lip sync unless you deliberately ask for audio only, with dynamic duration so the dub stretches or compresses to fit the original timing. That single action changes the economics of a launch: one shoot becomes every market on the supported list instead of one market per shoot. It also puts your presenter's face and mouth behind a claim in a language your team may not read, which is a different kind of risk from a bad render and deserves a different kind of control. The whole surface is asynchronous, and On Video Event delivers finished work through a real registered webhook rather than a polling loop, which keeps a six-market launch from turning into six timers.
Without FlowRunner
With FlowRunner
Use Case Scenarios
Six market launch from one master cut
A launch is scheduled across six markets with one English master. The agent validates each target against List Translation Languages, then submits Translate Video per market with lip sync and translated subtitles. On Video Event fires as each dub completes. The agent posts each finished cut into that market's channel in Microsoft Teams with the English source script beside it, and holds the campaign record in HubSpot at Pending Localization until every market has signed off. The shoot happened once. The approvals happen six times, on purpose.
Ad variants generated from the product record
A new SKU is published and needs social creative. The agent calls Generate AI Script with the product name and description, a 30 second runtime and a data-led style, then polls Get AI Script Result and receives several variants written in different styles. It posts them for a marketer to choose from rather than picking one itself, because the choice between a storytelling opener and a data-led pitch is a positioning decision. The chosen script goes into Create Avatar Video with a presenter resolved from List Public Avatars filtered by aspect ratio, style, age band and scene, and the finished MP4 URL is written back to the SKU record in Shopify.
Building and animating a house presenter
A brand wants a consistent presenter without contracting a person. The agent calls Generate Photo Avatar with the age band, style, background and appearance the brand guidelines specify, polls Get Photo Avatar Status until it reports success, then calls Add Motion To Photo Avatar and polls Get Motion Status until the animated preview is ready. Only then does the avatar become usable in video actions. Before that presenter appears in anything customer facing, the preview goes to the brand owner, because a face that will front every video the company publishes for the next year is a brand decision, not a rendering detail.
Human-in-Loop Highlight
Translate Video is the operation that earns a gate on this connector. It produces an AI voiceover in a target language and lip-syncs your presenter to it, so the finished cut shows a real, recognizable person appearing to say words that were generated, in a language the people approving the campaign frequently cannot read. The failure mode is not a bad render, which anyone can spot. It is a claim that is compliant in English and regulated in the target market, delivered in the spokesperson's own face. So the agent does not treat a completed dub as a finished asset. When On Video Event reports the German cut is ready, it routes to that market's reviewer with everything needed to judge it: "German dub ready for the Q3 launch. Source script attached, runtime 31 seconds against 29 in the master. You are the approver for DE. Confirm the voiceover matches the source claim and the wording clears local advertising rules, or send it back." Six markets, six approvals, one shoot. The agent handled every render, every webhook and every status check. It stopped at the six moments where being wrong is a regulatory problem rather than a production one.
Agent Capabilities
27 actionsAvatar Videos
2- Create Avatar Video Starts an asynchronous talking-avatar video from either a written script that JoggAI voices with text to speech, or an audio file you already have. Choose a public or custom avatar, the aspect ratio, and whether the avatar fills the frame, sits beside your content or floats picture in picture. Rendering takes a few minutes, so poll the status action or subscribe the trigger.
- Get Avatar Video Status Retrieves an avatar video by its ID, reporting whether it is pending, processing, completed or failed. Once complete the response carries the hosted MP4 URL, a cover image URL and the duration, so the links pass straight to a publishing step.
Template Videos
2- Create Template Video Renders a video from one of your templates by filling in its declared variables, which can be text, images, videos or scripts. Optionally override the presenting avatar, the voice, the background music and the caption setting, and disable automatic transitions and image movement when you need a static, brand-controlled look.
- Get Template Video Status Retrieves a template-based video by its ID with its current render status and, once complete, the hosted MP4 URL, cover image and duration.
Lip Sync and Dubbing
5- Create Lip Sync Video Re-times an existing video so the speaker's mouth matches a separate audio track, which is how narration gets swapped or a clip gets dubbed without reshooting. When the source video is shorter than the audio, the playback strategy controls whether the footage loops or ping-pongs to fill the time.
- Get Lip Sync Video Retrieves a lip sync task by ID with its processing status and, once complete, the hosted URL of the re-timed video.
- Translate Video Dubs a video into another language with an AI voiceover and optional translated subtitles, lip-syncing the speaker unless you choose to translate the audio only. Dynamic duration lets the dub stretch or compress so the timing still fits the footage. This is the action that puts generated words in a real presenter's mouth, so it is the one that belongs behind a native reviewer.
- Get Translation Status Retrieves a video translation task by ID with its current status and, once complete, the hosted URL of the dubbed video.
- List Translation Languages Returns the target languages JoggAI can dub a video into. The authoritative list to validate a market plan against before any translation work is submitted.
Scripts
2- Generate AI Script Writes marketing video scripts from product information, either an existing product ID or a name and description supplied inline. Pick the language, a target runtime of 15, 30 or 60 seconds, and a writing style ranging from a reassuring opener through storytelling narrative, a discovery angle, a data-led pitch, a top three reasons rundown or light marketing. Generation is asynchronous and produces several variants.
- Get AI Script Result Retrieves the outcome of a script generation task. When it completes, the response carries several variants written in different styles, so a flow can present them for a choice or feed the chosen text straight into Create Avatar Video.
Avatars and Voices
7- List Public Avatars Returns the public avatar library, usable immediately with no setup. Narrow by aspect ratio, professional or social style, gender, age band, scene setting and ethnicity to find a presenter that matches the brief. Each entry exposes the numeric avatar ID the video actions expect plus a preview image.
- List Custom Avatars Returns the instant avatars trained on your own footage, with their processing state so a flow can wait for a new avatar to finish before using it. Filter by name or by whether an avatar is processing, completed or failed.
- List Voices Returns the text-to-speech voice catalog covering 40+ languages, with a sample URL for each entry so a voice can be auditioned before it is used. Filter by language, gender, age band and use case. Each entry exposes the voice identifier the video actions expect.
- Generate Photo Avatar Creates a new presenter from a description rather than from footage, controlling age band, professional or social style, gender, ethnicity, background and appearance. The classic and modern models give distinctly different looks, and a reference image can anchor the result.
- Get Photo Avatar Status Checks whether a generated photo avatar is ready, reporting processing, success or failure with a message explaining any failure. Once successful, the avatar can be animated.
- Add Motion To Photo Avatar Animates a still photo avatar so it can speak in videos, turning a generated portrait into a usable presenter. The work runs in the background.
- Get Motion Status Checks whether a photo avatar has finished animating. Once complete the response carries the motion video URL and the source photo, and the avatar becomes available to the video actions.
Templates and Assets
5- List Templates Returns the stock template library filtered by aspect ratio. Templates are pre-designed video layouts whose variables get filled in with Create Template Video.
- List Custom Templates Returns the templates saved in your own workspace, optionally filtered by aspect ratio.
- Get Custom Template Retrieves one of your custom templates with the full list of variables it declares, including each variable's name and type. Call this before Create Template Video so the variables you send match exactly what the template expects.
- List Background Music Returns the background music tracks available for template videos, paginated. Each entry exposes the music ID that Create Template Video accepts.
- Create Asset Upload URL Requests a signed cloud storage URL for uploading an image, video or audio file into your workspace. JoggAI does not accept bytes directly: it returns a signed URL that your own step then uploads to, after which the asset can be referenced in video requests.
Account and Webhooks
4- Get User Info Returns the account behind the configured API key. Used as a connection check at the start of a flow, or to confirm which workspace a video will be created in.
- Get Remaining Quota Reports the API credits left on the account. Used to guard a batch of renders so a run does not stop part way through.
- List Webhook Events Returns every webhook event type JoggAI can notify on, covering video, avatar, image, motion and script generation successes and failures. Used to confirm the exact event names available on your account before subscribing.
- List Webhook Endpoints Returns the webhook endpoints registered on the account with their URLs, enabled state and subscribed events. JoggAI allows up to twenty endpoints per account, so this is how you audit what is registered, including endpoints the trigger created automatically.
Triggers
1 triggersEvent Triggers
1- On Video Event Fires when a JoggAI job reaches a terminal state, delivering the finished video or avatar without any polling. Pick the events to subscribe to across avatar, template, product and translated video successes and failures, plus avatar, photo avatar, image, motion and script events. The extension registers a real webhook endpoint on your account when the flow is deployed and removes it when the flow is undeployed. JoggAI expects a 200 response within five seconds and retries otherwise, so keep the receiving step fast and do the real work after acknowledgement.
Frequently Asked Questions
What can FlowRunner do with JoggAI?
FlowRunner agents can run Create Avatar Video, Get Avatar Video Status, and Create Template Video in JoggAI, plus 24 more actions.
Does connecting JoggAI to FlowRunner require OAuth?
No. JoggAI connects to FlowRunner with an API key, no OAuth flow required.
Can JoggAI trigger a FlowRunner workflow automatically?
Yes. JoggAI supports 1 trigger that can start a FlowRunner workflow automatically.
Start building with JoggAI
$100 in credits. No card required. Connect in minutes.