ceven

Gan.AI

Converts text into natural speech and generates personalized avatar videos to automate customer outreach and training content at scale.

Try Gan.AI in Ceven

Ask Ceven anything
Standard

Why use Ceven?

  1. AI native Gan.AI integration

    • Describe the outcome and Ceven picks the right Gan.AI calls, fills the parameters, and checks the result.
    • Structured, agent friendly tool schemas so each call runs reliably instead of by guesswork.
    • Rich coverage for reading, writing, and querying your Gan.AI data, across all 8 of its actions.

  2. Managed auth

    • Built in OAuth with automatic token refresh and rotation.
    • One place to manage, scope, and revoke Gan.AI access.
    • Per user and per environment credentials instead of shared keys.

  3. Agent optimized design

    • Actions are tuned from real success and error rates so reliability climbs over time.
    • Full execution logs so you always know what ran in Gan.AI, when, and on whose behalf.
    • The agent pauses and asks when Gan.AI is unclear instead of plowing ahead.

  4. Enterprise grade security

    • Fine grained access so you control which agents and people can reach Gan.AI.
    • Least privilege by default, read scopes first and only the writes a workflow needs.
    • A full audit trail of every Gan.AI action to support review and sign off.

Supported tools

Every action Ceven's agents can run on Gan.AI, and when to use it.

Convert text to speech
Use this when you need to turn a text string into an audio file synchronously using a specific voice ID.
Get avatar video details
Pull metadata and the final video URL for a specific inference ID to check if the render is finished.
Get photo avatar details
Retrieve the metadata and image URL for a specific photo avatar by its unique ID.
Get photo avatar inference details
Fetch the current status and processing details for a photo avatar video request using an inference ID.
List avatar video inferences
Pull a list of generated avatar videos filtered by status, date, or avatar ID for batch reporting.
List photo avatar inferences
Retrieve a collection of photo avatar video results, useful for auditing recent generation requests.
List photo avatars
Query the library of available photo avatars using filters like creation date or published status.
Authenticate user
Use this to login with email and password to retrieve the access tokens needed for all other calls.
Refresh access token
Mint a new dynamic access token using existing authorization to keep a long running workflow active.
Create voice clip
Generate a short audio snippet from text for use in notifications or simple voice alerts.
Check render status
Verify if a specific video ID has moved from processing to completed status.
Filter avatars by date
Search for photo avatars created after a specific timestamp to find the newest assets.
GAN.AI Login
Tool to authenticate a user and retrieve tokens. - If email/password are provided: performs classic login (POST /users/login). - Otherwise: mints a dynamic access token using existing Authorization (POST /users/create_dynamic_token_v2).
GAN AI Text to Speech
Tool to convert text to speech synchronously. Use after selecting a voice ID via Get Voices tool.
Get Avatar Video Inference Details
Tool to retrieve detailed status and metadata for a specific avatar video inference. Use when you have an inference_id and need to check its processing status and access video URLs.
Login
Tool to authenticate a user and retrieve access and refresh tokens. Use when you need to login before calling other GAN.AI API actions.

16 actions · scroll to see them all

Frequently asked questions

Ceven implements a polling logic known as a wait state. When the agent triggers a video inference, Gan.AI returns an inference ID rather than a video file immediately. The Ceven agent stores this ID and periodically calls the Get avatar video inference details action. Once the status field changes from processing to completed, the agent retrieves the final URL and proceeds to the next step in your workflow, such as sending the video in an email or posting it to a social media platform. This ensures that your customers never receive a broken link or a pending request notification.
You must use a valid voice ID that exists within your Gan.AI account. To find the right voice, you can use the list voices functionality to see which personas are available for your specific plan. Once you identify the voice ID that matches your brand tone, you provide that ID to the Ceven agent. The agent then passes this ID in the API request to ensure the output audio sounds exactly as intended. If you have cloned a custom voice, that specific ID should be used for all personalized outreach workflows.
Ceven monitors the status of every request. If the Gan.AI API returns a failed status for a specific inference ID, the agent triggers an error handler. Depending on your workflow settings, it can either attempt to regenerate the video once or send a notification to your team via Slack or email alerting you that the video failed to render. This prevents the workflow from hanging indefinitely and ensures that you are aware of any assets that did not generate correctly due to input errors or system issues.
Ceven is subject to the rate limits and credit quotas of your Gan.AI subscription tier. One specific quirk of the Gan.AI API is that high resolution video renders are gated by a concurrency limit, meaning you can only have a few videos processing at the exact same time. If you trigger a massive batch of videos, Ceven will queue the requests and space them out to avoid receiving a rate limit error. We recommend checking your Gan.AI dashboard to see your current credit balance to avoid workflow interruptions.
No, Ceven does not store the actual video or audio files. We only store the metadata and the temporary URLs provided by Gan.AI. The files themselves reside on Gan.AI servers. When your workflow needs to deliver a video, the agent fetches the current URL and passes it to the destination tool. If you need a permanent archive of these videos, we recommend adding a step in your workflow to upload the file from the Gan.AI URL to your own cloud storage like Amazon S3 or Google Drive.
Ceven uses a secure token exchange system. You provide your Gan.AI credentials via the secure connection portal, and the agent uses the login tool to retrieve an access token and a refresh token. These tokens are encrypted at rest. When a workflow runs, Ceven uses the access token to make requests. If the token has expired, the agent automatically uses the refresh token to obtain a new one without requiring you to re authenticate manually, ensuring that your automated video pipelines run without human intervention.
The current integration focuses on the inference and retrieval of avatars. You typically create and train your photo avatars within the Gan.AI web dashboard to ensure the quality and lighting are correct. Once the avatar is published and assigned an ID, you can use Ceven to drive the generation of infinite videos using that specific avatar. This separation ensures that the high touch process of avatar creation is handled by a human while the high volume process of video production is handled by the agent.
Yes, the language is determined by the voice ID you select. Gan.AI supports multiple languages and accents across its voice library. When you configure your Ceven workflow, simply ensure you are calling a voice ID that corresponds to the target language. The agent will send the text to the API and Gan.AI will handle the phonetic synthesis based on that voice profile. For multi language campaigns, you can build a logic branch in Ceven that selects a different voice ID based on the customer language field in your CRM.

Alternatives to Gan.AI

Other tools that solve a similar problem. Ceven supports these too, so you can switch or run more than one at once.

HeyGen logoHeyGenSynthesia logoSynthesiaElevenLabs logoElevenLabs

Try Ceven on your stack

Plug Ceven on top of the tools you already run. Connect Gan.AI and the rest of your stack, describe the outcome, and its agents handle the work end to end, days of it in minutes.

Get started for free