Frequently Asked Questions

Descrideo is an AI-powered video description service that automatically analyzes your videos using visual frames, audio transcription, or both to generate detailed, accurate text descriptions. Choose from three generation modes: vision-only, combined vision + audio, or audio-only. Our technology is perfect for accessibility, SEO, and content management.

Our service follows a simple process: First, you upload your video to our secure platform. Then, depending on your chosen generation mode, our AI extracts key frames, transcribes audio segments, or both. Advanced computer vision and speech recognition models analyze the content. Finally, you receive detailed descriptions via webhook or in your dashboard. The entire process is automated and typically completes within seconds to minutes depending on video length and generation mode.

Descrideo offers three generation modes: Vision (default) analyzes extracted video frames to understand visual content. Vision + Audio combines frame analysis with audio transcription for the most comprehensive descriptions — ideal for vlogs, reviews, and presentations. Audio Only transcribes speech without frame extraction, perfect for podcasts, interviews, and lectures. Vision mode is available on all plans including free. Audio modes require a paid plan.

When you enable audio transcription (vision_audio or audio mode), our system extracts audio segments from your video and transcribes them using advanced speech recognition. You can configure the number of segments (10, 20, or 30) and segment duration (5-60 seconds each). The transcribed text is then combined with visual analysis (in vision_audio mode) or used as the sole input (in audio mode) to generate descriptions. Audio transcription is billed as an add-on based on the total sampled audio duration.

Descrideo supports all major video formats including MP4, MOV, AVI, MKV, WebM, and more. Our system automatically handles video conversion and optimization for AI analysis. Maximum file size varies by plan.

Yes, security is our top priority. All videos are stored using encrypted S3-compatible storage with access controls. We use HMAC-verified webhooks for secure communication, and all data transmission uses HTTPS encryption. You can delete your videos and associated data at any time.

Descrideo can generate video descriptions in multiple languages. Our AI is capable of producing descriptions in English, Spanish, French, German, and many other languages. You can specify your preferred output language when creating a description job.

Descrideo offers a robust API and webhook system for seamless integration. You can send video description requests via our REST API and receive results through webhooks. All webhook communications are secured with HMAC signatures for verification. Check our documentation for detailed integration guides.

We offer flexible pricing plans to suit different needs, from individual creators to enterprise solutions. Create a free account to get started and explore our features. Contact our sales team at contact@descrideo.com for custom enterprise pricing.

Audio transcription is an add-on to the base job cost. You pay the base token cost for each successful job, plus an additional cost based on the amount of audio sampled (calculated per 10-second increments). The exact cost is displayed before job creation and confirmed in the webhook billing payload. Audio modes (vision_audio and audio) are available on all paid plans. The free Demo plan is vision-only.

Our AI achieves high accuracy by analyzing multiple frames and optionally transcribing audio from your video. The combined vision + audio mode produces the richest descriptions by capturing both what's shown and what's said. Audio-only mode excels for podcasts, interviews, and lectures where speech carries the primary information. Accuracy can vary based on content complexity and audio quality.

Descrideo is designed with accessibility in mind. Our AI-generated descriptions can be used as audio-description scripts (for narration), as text alternatives/media alternatives, and as supporting context alongside captions. The combined vision + audio mode provides the most complete accessibility coverage by capturing both visual and spoken content. Final accessibility compliance depends on your implementation and review process.

Email us at contact@descrideo.com. We typically respond within 24-48 business hours. You can also check our FAQ and documentation for quick answers to common questions.