Gan.AI

@Gan.AI list all my published photo avatars created in the last 30 days, then show me the processing status for each one

Found 4 published photo avatars from the last 30 days: 1. **Corporate Headshot** (ID: av_8x3k) — Status: completed, created Jan 15. Image URL available. 2. **Product Demo Avatar** (ID: av_9m2p) — Status: completed, created Jan 18. Image URL available. 3. **Training Video Host** (ID: av_1n7q) — Status: processing, created Jan 22. Estimated completion in 2-3 minutes. 4. **Customer Support Rep** (ID: av_4k8s) — Status: failed, created Jan 20. Error: insufficient image resolution. This example shows how the MCP retrieves avatar metadata including processing state and access URLs when rendering is complete.

Demonstrates the List Photo Avatars tool with date filtering and status checking. Requires valid API key with read access to avatar resources. Processing times vary by queue load; failed avatars often need re-upload with corrected source material. The MCP returns paginated results—large avatar libraries may need multiple queries.

@Gan.AI convert this script to speech using a professional female voice: 'Welcome to our Q1 product update. Today we're announcing three major features that will transform how your team collaborates.'

Speech synthesis complete. Generated a 7-second audio file using voice ID `voice_f3_pro_en` (Professional Female, English). Audio URL: `https://gan.ai/audio/syn_9x4k2p.mp3` (expires in 24 hours) This example illustrates synchronous text-to-speech conversion. The MCP selected a voice matching your 'professional female' description from the available voice library, processed the text, and returned a playable audio file. You'd typically use Get Voices first to browse options, then reference a specific voice_id for consistent branding across multiple scripts.

Showcases the Text-to-Speech tool for rapid audio generation. Requires prior voice selection via Get Voices—this example assumes you've already identified your preferred voice_id. Generated audio URLs are temporary; download immediately if you need permanent storage. Synchronous processing means longer scripts may hit timeout limits; consider breaking text into smaller chunks for narration over 30 seconds.

@Gan.AI check the status of my avatar video inference av_inf_7k3m and if it's ready, summarize what the video contains based on the title and metadata

Avatar video inference **av_inf_7k3m** is complete. **Title:** "Sales Pitch - Enterprise Plan Overview" **Status:** completed **Duration:** 1:42 **Video URL:** `https://gan.ai/videos/av_inf_7k3m.mp4` **Created:** Jan 23, 2025 at 3:14 PM Based on the title, this appears to be a generated avatar video presenting your enterprise product tier. The 1:42 runtime suggests a concise pitch covering 2-3 key features. This example shows how the MCP retrieves inference details including processing state and playback URLs once rendering finishes.

Highlights the Get Avatar Video Inference Details tool for tracking video generation jobs. Avatar video rendering is asynchronous—initial requests return an inference_id you poll until status becomes 'completed'. Processing time depends on video length and queue depth (typically 2-10 minutes). Failed inferences often indicate script/avatar mismatches; check error messages in the metadata. Video URLs expire after 30 days unless downloaded.

When Gan.AI beats hiring voice actors for multi-market launches

A 6-person SaaS team shipping to three regions needs product walkthrough videos in English, Spanish, and German. Gan.AI's text-to-speech and avatar video tools let them script once, generate three localized versions in under an hour, and update the script when features change without re-recording. The MCP's inference-listing tools make it easy to track which versions are rendering and grab URLs for embedding. This works if your script is under 5 minutes and you're okay with synthetic voices—longer videos or high-stakes sales demos still need human talent. If you're launching fast and iterating often, Gan.AI keeps video content in sync with your roadmap without ballooning production costs.

Scaling video FAQs without a studio or editing team

A 12-person support org gets the same 20 questions every week. Writing help docs is fast, but users want video. Gan.AI's photo avatar and text-to-speech tools turn existing support articles into short explainer videos—no camera, no editing software, no voice actor contracts. The MCP's list-inferences tools let the support lead batch-generate 15 videos on Monday, check status Tuesday, and publish URLs to the knowledge base by Wednesday. This breaks down if your FAQs need screen recordings or live product demos; Gan.AI handles talking-head content, not screencasts. If your support backlog is text-heavy and your team has zero video production bandwidth, this MCP turns written answers into watchable content in a single afternoon.

When avatar videos make cold outreach feel less cold

A 3-person sales team prospecting into enterprise accounts wants to send personalized video intros instead of generic emails. Gan.AI's avatar video inference tools let them script a template, swap in the prospect's name and pain point, and generate 50 unique videos in a batch. The MCP's inference-listing and detail-retrieval tools make it scriptable—pull the video URLs, drop them into the CRM, and track which prospects watched. This only works if your pitch is under 90 seconds and your ICP tolerates synthetic avatars; some buyers find them off-putting. If you're testing video outreach and don't want to record 50 takes yourself, Gan.AI gets you to proof-of-concept in a day without hiring a videographer.