daydreams

What is Daydream AI?

Daydream AI is a platform for creating long-form videos from stories, novels, or creative ideas; the current web interface uses the MagicLight brand, while some application pages still retain the name “Daydream”.

This product integrates tasks such as script breakdown, storyboard creation, character development, image generation, dynamic transitions, voiceover recording, and final video output into a single workflow, and is designed primarily for creators of short dramas, comic series, novel-based tweets, and story videos.

Brand and operational information

• The website identifier has been changed to MagicLight, but the page title, as well as the entries in mobile app stores, still show “Daydream AI”.
• The current website registration information and the field indicating the service provider in the app store point to Shenyang Feihang Network Technology Co., Ltd.
• The developer name in the app store remains Guangmo Technology (Shenzhen) Co., Ltd.; however, the labels displayed on different platforms are not exactly the same.
• Subscriptions, invoices, refunds, and data responsibilities shall be determined in accordance with the actual account holder and the entity listed on the order and settlement pages.

Main functions

Turning stories and copy into videos

• Users can enter excerpts from novels, story outlines, tweet texts, or creative descriptions, and the system will organize them into video projects that can be produced.
• Smart scripts can identify characters, scenes, dialogue, and narrative points, and then break down longer texts into camera tasks.
• The intelligent scripts associated with member benefits support a duration of up to about 50 minutes and a length of up to 6,000 Chinese characters; however, these limits may vary depending on the account, the entry point, and the type of task.
• Long texts do not guarantee a complete and error-free final product; users still need to check for omissions, causal relationships, character motivations, and pacing.

AI scripts and storyboards

• The system can convert text into a storyboard, assigning characters, shot angles, actions, settings, and narration for each scene.
• Creators can first review the script and the structure of the shots, and then create the individual elements one by one, which reduces the costs associated with having to redo work after the entire video has been produced.
• The script skills entry is used to reuse specific narratives or to rewrite tasks, and it is suitable for creating a content flow that is updated continuously.
• Automatic scene breakdown may overlook foreshadows, emotional progression, and spatial continuity; important shots should be rearranged manually by the creator.

Role creation and role consistency

• Users can create character avatars using photos or descriptions, and reuse the same character in multiple shots.
• The character library provides community characters, visual previews, usage counts, and creator information; it also allows users to create their own character assets.
• Fixed characters can reduce visual drift across consecutive shots, but clothing, age, facial features, posture, and lighting can still change.
• Making a character publicly available may allow other users to use that avatar; permission for the use of the portrait must be obtained before uploading a real-person photo.

Visual style and image generation

• The visual style library covers various creative genres such as 3D cartoons, urban romance, wuxia-style stories with a traditional atmosphere, realistic styles, and mystery/thriller narratives.
• Users can select a style for a project or shot, and then combine it with descriptions of characters, scenes, and actions to create static images.
• Style templates help maintain the visual tone of a series of works, but different models and complex prompts can still lead to variations in the painting style.
• When dealing with living artists, brand visuals, or protected characters, commercial usage rights should not be determined solely based on style options.

Convert static images into animated videos

• Static images that are generated or uploaded can be further converted into animated sequences; the system adds actions of characters, camera movements, and changes in the environment as indicated.
• Pro dynamic conversion is part of the member benefits; the available models, resolutions, duration, and amount of Dream Coins required are determined based on the task preview.
• Complex limbs, multiple character interactions, rapid movements, and small objects tend to experience deformation, interference, or discontinuous motion.
• Creators can shorten individual shots, simplify actions, and produce them in segments, then improve stability through editing.

Dubbing and voice cloning

• The platform allows different voices to be assigned to dialogues and narration, making it suitable for stories with multiple characters and explanatory videos.
• Member benefits include voice cloning, which allows creating more personalized voice actors using available voice samples.
• Voice cloning requires explicit permission from the person whose voice is being cloned, and it must not be used for impersonation, fraud, or creating unauthorized statements.
• Dialects, emotions, accents, numbers, and proper nouns may be pronounced incorrectly; it is necessary to listen to each sentence before publishing and make the necessary corrections.

Video editing and final output

• Users can adjust text, visuals, animated effects, and voiceovers at the clip level, and then combine multiple clips to create a complete video.
• The paid features support 1080P videos, and the descriptions of related applications also mention MP4 and MOV output formats.
• Watermark-free and high-definition exports should be determined based on the membership level purchased and the privileges associated with the current order; free accounts should not have the same rights by default.
• For entire videos, it is still necessary to check the subtitles, audio-video synchronization, volume level, aspect ratio, transitions, and the guidelines for publishing on various platforms.

Asset store and community resources

• The platform features an asset store, a character library, and access to visual style options, allowing users to reuse existing creative assets in their projects.
• The public character page displays the character’s image, the creator, and its popularity, making it useful for finding inspiration for creation.
• The fact that a community is available for use does not mean that commercial rights to character prototypes, celebrity portraits, film or television characters, or trademarks can be obtained.
• Enterprises and commercial projects should keep records of authorizations, and verify the rights of third parties for each item before making any releases.

Input and output

Complete usage process

1. Prepare the stories, copy, photos, audio, and reference materials needed for usage rights, and determine the video aspect ratio and target duration.
2. You can log in using a mobile phone verification code or password; if you don’t have a registered mobile number, an account may be created automatically during the login process.
3. Enter AI video creation, paste text or fill in a story outline, and select the appropriate script processing method.
4. Review the characters, scenes, and storyboards, and first correct the character relationships, the order of the shots, as well as the dialogue and narration.
5. Create a custom character or select an avatar from the character library to establish a fixed appearance and style for the main character.
6. Generate images based on the camera footage; if there is any distortion in the clothing, faces, or background, describe them again or redo the process promptly.
7. Select the dynamic model and motion cues to convert static images into short clips, and check the amount of Dream Coins consumed before submitting.
8. To configure multi-character voice acting, authorization must be obtained and compliant samples uploaded first if voice cloning is required.
9. Examine each frame in detail to check the actions, lip movements, subtitles, audio, and transitions, then combine them into a complete project.
10. Export videos based on account privileges, and conduct a review of facts, copyrights, portraits, trademarks, and platform rules before publishing.

Member prices and Dream Coins

The different interfaces for premium members provide varying estimates of the number of minutes of video that can be generated; therefore, it is not appropriate to consider any specific number of minutes as a fixed commitment.

The original price, discounted price, in-app purchase amounts, and international plan options displayed on the page may vary depending on the region, sales channel, promotions, taxes, and settlement period; therefore, it is necessary to refer to the actual settlement page and account details before making a purchase.

How to understand the consumption of dream coins

• Dream Coins are the unit used to measure the amount of resources required for generating tasks; different image models, video models, as well as different duration and quality settings may require different numbers of these coins.
• The number of images and the duration in minutes for the package are estimates based on specific usage levels; not all tasks are deducted in the same proportion.
• The rules regarding failure in generation, retry attempts, cancellation, upgrades, and the carryover of remaining Dream Coins are not fully disclosed; they should be confirmed before submitting a task and making a payment.
• App stores offer various in-app purchase options with different prices, but the names of these options do not clearly indicate the benefits they entail; therefore, it is not possible to determine the package in question solely based on the price.

Platform support

The current application offers a Simplified Chinese interface; it has not yet been confirmed whether the website will function properly in all mobile browsers, nor the status of a native Android application.

API, SDK, and open-source status

• The platform’s web pages and applications call internal interfaces related to orders, members, and projects; however, these internal interfaces are not the same as the APIs made available to developers.
• At present, there are no available public API documents, options for requesting developer keys, an official SDK, or clear pricing information for commercial interfaces.
• At present, no official open-source repository or open-source license for the product itself has been identified; therefore, Daydream AI cannot be considered an open-source video generation tool.
• The enterprise solution mentions customizable functions; if system integration is required, it is necessary to consult the sales team regarding the scope of the interfaces, service levels, data processing, and costs.

Which users are it suitable for

• Creators who wish to quickly turn novels, tweets, or story outlines into manga and narrative videos.
• A short-form drama team that needs to continuously maintain the character images, visual style, and voiceovers for multiple characters.
• Individual users who do not have a complete animation production team, but are willing to review each frame and make corrections to the AI-generated content.
• Workshops that are capable of producing story videos in bulk, as well as assessing the costs related to Dream Coins, copyright issues, and content moderation.
• Enterprise clients who wish to establish custom generation processes, priority queues, or team quotas.

Product advantages

• It transforms text into a cohesive long-video process that includes scripts, storyboards, characters, animations, and voiceovers.
• The role library and visual style library reduce the effort required for repetitive configuration of series content.
• Lens-level generation and editing allow users to make partial changes, without having to rebuild the entire video each time.
• Members provide conversions based on Dream Coins, the number of images, and the duration of certain videos, which helps in estimating the scale of production.
• Websites and mobile applications cover different creative scenarios and offer support for high-resolution output.

Capacity limits and usage restrictions

• Character consistency, camera continuity, lip synchronization, hand movements, and complex actions cannot be guaranteed to be correct every time.
• Long-form projects exacerbate issues such as character drift, scene transitions, narrative omissions, and audio-video synchronization problems.
• The generation speed is influenced by the task length, the model used, the level of clarity, the queue status, and the priority of the account.
• The number of minutes indicated by members is an estimate of usage, and it cannot be regarded as a fixed commitment regarding the quantity or quality of the output.
• It is possible that third-party models are not available, and the amount of resource consumed per use can vary; therefore, it is necessary to check the real-time prompts when selecting a task.
• The results generated by the platform may contain factual errors, similar images, visual defects, and content that is not suitable for publication.
• Invitation rewards and promotional benefits may vary depending on the campaign or account, and they are not fixed free allowances available to all users.

Privacy and data security

• Logging in allows access to account information such as phone numbers; during the creation process, text, images, audio, video, and operation records are also processed.
• The iOS Privacy Report lists names and phone numbers that may be linked to an individual’s identity, as well as various types of usage data and user content that are not tied to any identity.
• Application privacy information is submitted by the developers, and it cannot replace the user’s own verification of the actual policies, permissions, and online activities.
• The public page does not provide complete information regarding the retention period for prompts and materials, their intended uses for training, the deadline for deletion, as well as matters related to cross-border transfers and sub-processors.
• Commercial secrets, unpublished scripts, identification documents, child-related information, or unauthorized real-person images and audio should not be uploaded.
• Before proceeding with group purchasing, it is necessary to obtain written confirmation regarding data isolation, access control, backup procedures, deletion processes, incident notification protocols, and account cancellation procedures.

Copyright, portrait rights, and commercial usage notes

• Users must possess the necessary rights to the novels, images, music, fonts, characters, sounds, and trademarks they input.
• When the character library includes recognizable real people or well-known fictional characters, the platform can use them, but this does not mean they can be utilized in advertisements or commercial releases.
• Voice cloning involves risks related to personality and biological characteristics; it is necessary to keep track of the scope of authorization, its duration, its intended use, and the procedures for revoking it.
• The commercial use or watermark-free features included in the paid plans do not automatically remove third-party copyright, portrait rights, trademark rights, and model restrictions.
• Before publishing to short videos, advertisements, films, or paid courses, it is necessary to check the AI labeling rules and guidelines regarding synthetic content on the target platform.
• At present, there are no uniformly established deadlines and conditions for refunds; refunds for subscriptions, Dream Coins, and generated tasks shall be governed by the rules applicable to those orders.

Frequently Asked Questions

What is the relationship between Daydream AI and MagicLight?

The current website primarily uses the MagicLight brand, but the website title, mobile applications, and store listings still retain the White Daydream AI name; this can be seen as different representations of the same product during a brand transition.

Can Daydream AI convert an entire novel into a video directly?

It can break down longer texts into scripts and storyboards, and then generate visuals and voiceovers, but the length of the input text, the duration of the project, the available quota, and the quality of the content are still subject to restrictions imposed by the membership level and the model used.

How much content can be generated with the free version?

According to the details of the current package, the free version provides 500 DreamCoins at one time; the number of images or videos that can be generated depends on the model chosen, as well as on the duration and quality.

Why might the member price be different from that in the app store?

Websites, mobile devices, region, taxes and fees, events, and payment cycles can affect the price and the benefits offered; before making a purchase, it is necessary to check the package name, renewal options, and the amount of Dream Coins on the payment page.

How many minutes of video can a Gold member create?

There are discrepancies in the conversion of minutes across different interfaces; therefore, it is more appropriate to estimate costs based on 25,000 DreamCoins per month and the actual amount consumed as shown before submitting a task.

Can the generated video be used for commercial purposes?

It is necessary to meet the requirements of the purchased package, as well as those related to the models and materials used; furthermore, it is essential to ensure that any third-party rights related to novels, characters, portraits, voices, music, fonts, and trademarks have been properly authorized.

Does Daydream AI have a public API or an open-source version?

At present, there are no confirmed public developer APIs, official SDKs, open-source repositories for the product, or open-source licenses; therefore, the requests made within the website cannot be used as public interfaces for external integration.

What should be kept in mind when using voice cloning?

Clear consent from the owner of the voice must be obtained, and its use must be restricted; it cannot be used to impersonate others, engage in fraud, or create unauthorized sensitive content.

How to add a FAQ Schema

The page already employs a visible Q&A format in which the question title is followed directly by the corresponding answer paragraph; the site template can read these Q&A pairs on the server side to generate structured FAQ data.

Structured data must correspond word for word to the content visible on the page; it should not include hidden questions, price commitments, ratings, links, or any benefits that cannot be confirmed at the moment.

Summary

Daydream AI is suitable for converting story texts into long-form narrative videos with stable character portrayals, while the MagicLight workflow covers script development, storyboarding, visuals, animation, voiceovers, and export.

When selecting a package, it is important to compare the Dream Coins required with the actual resource consumption, and to complete the review of characters, portraits, voices, materials, and platform rules before the commercial release.