What is OhYesAI?
OhYesAI is an AI creation platform that transforms songs into complete visual narratives and music videos.
It integrates rhythm analysis, storyboard planning, shot creation, individual shot editing, and final composition into a single process.
Main functions
- Upload an MP3 and analyze its rhythm and song sections.
- Use the platform’s built-in AI features to create original music.
- Generate a reference image of the main subject based on the characters, scene, and props.
- Automatic generation of a storyboard script corresponding to the entire song.
- Generate videos frame by frame, allowing for individual review and modification.
- Adjust the lens, style, and rhythm using natural language.
- Combine all the shots into a MV of up to five minutes in length.
- Switch between different video generation models.
Five-step production process
| Steps | Main tasks | Key points of inspection |
|---|---|---|
| Upload music | Upload an MP3 or use AI music | File quality and music copyright |
| Confirm the entity | Define characters, environment, and props. | Reference diagram: Authorization and consistency |
| Confirm storyboarding | Review the script for AI-generated scenes | Stories, rhythm, and number of shots |
| Confirm video | Frame-by-frame generation and refinement | Characters, actions, and continuity of the visuals |
| Merge MVs | Combining music with all the footage | Beat marks, subtitles, watermarks, and export |
Music rhythm analysis
The platform analyzes BPM, drum beats, drops, and song sections to plan the shots and transitions.
- Identify the intro, verse, chorus, interlude, and outro.
- Adjust the timing of screen transitions based on the drum beats.
- Enhance camera changes at the climax and drop points.
- Give electronic, rap, and ambient music different rhythms.
- Even after automatic analysis, the creator should still check the effect of the key moments.
Automatic storyboard creation and complete narrative
OhYesAI not only generates individual segments but also attempts to create a cohesive visual narrative around the entire song.
- Plan the shots based on the lyrics, mood, and song structure.
- Assign scenes, shot sizes, and camera movements to different paragraphs.
- Maintain continuity between characters and props through the main reference diagram.
- It supports redoing each frame individually, without the need to regenerate the MV entirely.
- A complete video of up to five minutes can be output.
Dialogic editing
Users can modify individual shots using natural language, without having to learn complex timeline parameters.
- The location or weather conditions under which a certain lens is replaced.
- Adjust the tone, style, and overall atmosphere.
- Specify camera angles such as low-angle shots, high-angle shots, and push-in shots.
- Modify character movements, clothing, and props.
- Rearrange the correspondence between transitions and musical climaxes.
Consistency between reference diagrams and characters
The official FAQ states that one to seven reference images can be uploaded to help identify the main visual element.
| Reference content | Main function | Risk |
|---|---|---|
| Characters | Maintain the face and the appearance of the character. | Right to portrait and deepfake technology |
| Clothing | Maintain the shape and color | Brand logos and design copyrights |
| Environment | Unify the worldview and scenario | Location and material licensing |
| Props | Maintain the appearance of the items | Trademark and product accuracy |
| Visual style | Control of color and visual language | Avoid excessive imitation of living artists. |
Referencing images can improve consistency, but it cannot guarantee that the face, clothing, and props are exactly the same in every frame.
Supported video models
The official website currently displays a variety of models, allowing users to choose based on image quality, speed, and cost.
| Model | Official website positioning | Suitable direction |
|---|---|---|
| Vidu Q2 | Taking into account both cost and lighting effects. | Most conventional MV styles |
| Kling V3 Omni Pro | Emphasize camera movement and large-scale actions. | Realistic characters and dynamic scenes |
| Seedance 2.0 Fast | Quick preview and iteration | Verify the storyboard and creative direction |
| Seedance 2.0 | Emphasis on details and image quality | Lenses that achieve a film-like quality |
The list of models, available resolutions, and points consumption may change; the information displayed on the creation interface should be taken as reference.
Visual style
- A style reminiscent of healing animations and youth movies.
- Oil painting, hand-drawn art, and ink-based visual arts.
- Cyberpunk and neon cities.
- Surrealism, dream cores, and abstract narrative.
- Realistic films and live music genres.
- When using style references, avoid infringing upon others' work through imitation.
Which users are it suitable for
- Independent musicians and bands.
- AI music and original song creators.
- Operators of short-video and music channels.
- Record labels and promotion teams that need to visualize songs.
- A studio that creates MVs for clients.
- Individual users who lack experience in professional video editing.
OhYesAI usage guide
- Register an account and receive free trial points.
- Upload an MP3 file for which you have usage rights, or create original music.
- Fill in the song’s theme, story, target audience, and visual direction.
- Upload one to seven compliant reference images to lock in the subject.
- Check the settings for the AI-generated characters, environments, and props.
- Review the correspondence between the storyboard and the various sections of the song.
- Choose a generation model that fits the budget and the desired output quality.
- Generate frame by frame and check the characters, actions, and continuity.
- Use natural language to modify specific shots that are not satisfactory.
- Preview the synthesized result and check for timing issues, subtitles, and volume.
- Confirm the package, commercial rights, and material licensing before exporting.
Price and package comparison
The following are the information on RMB packages as displayed on the official website on August 30, 2026.
| Plan | Monthly payment | Annual equivalent | Integration | Estimated output |
|---|---|---|---|---|
| Free trial | 0 yuan | Not applicable | 1500 points awarded upon registration. | The film is formed in about 15 to 30 seconds. |
| Basic version | 78 yuan per month | 65 yuan per month | 7800 points per month | Approximately 2 to 3 60-second MVs |
| Standard version | 160 yuan per month | 130 yuan/month | 18,000 points per month | Approximately 5 to 6 60-second MVs |
| Professional Edition | 320 yuan/month | 260 yuan per month | 40,000 points per month | Approximately 12 to 13 60-second MVs |
The page states that a 20% discount is available for annual payments; the final amount to be paid, as well as any taxes and fees, are subject to the details on the settlement page.
Differences between the free version and the paid version
| Project | Free trial | Paid plans |
|---|---|---|
| Length of the sequence | About 15 to 30 seconds | Supports full video recordings of up to 5 minutes in length. |
| watermark | There is a watermark. | Export without watermarks available |
| Commercial use | Not supported | Includes commercial licensing. |
| Integration | A free 1500 is given upon registration. | 7,800 to 40,000 per month |
| Complete process | Available for experience | Use continuously as part of a package |
Points package
The official website states that the point packages offer between 4,800 and 32,000 points, but they must be purchased as part of a paid plan.
- The validity period of the points package is twelve months.
- The home page does not list the specific prices for each tier of point packages.
- Before making a purchase, it is necessary to check the rules regarding the use of points after the membership expires.
- Different models and lens settings may consume different amounts of points.
- To determine whether points are refunded for failed tasks, it is necessary to check the real-time protocol.
Commercial use and copyright
The Basic, Standard, and Professional versions all indicate options for watermark removal and commercial licensing.
Platform authorization covers only the portions it is able to authorize; it cannot replace the user’s legitimate rights over the input materials.
- Verify the copyright of the uploaded songs, accompaniments, and lyrics.
- Obtain permission for the use of real-person photos.
- It is not allowed to create misleading MVs using unauthorized images.
- The brands, characters, and artistic works referenced in the diagram must be legal.
- Comply with the rules of music and video platforms before publishing.
- Commercial projects should keep order records, as well as records of generation and authorization.
Uploading music and privacy
The official FAQ states that music uploaded is intended solely for creating MVs, and not for model training or distribution to external parties.
Users should still read the privacy policy and avoid uploading unreleased master tapes that are subject to confidentiality agreements.
- Keep the original local files and project backups before uploading.
- Do not include ID numbers or contact information in the file name.
- Team works require the consent of all co-creators.
- After the project is completed, check the file deletion and saving mechanisms.
- Sensitive business projects should first have their risks associated with cloud processing assessed.
GitHub and open source
No official, complete open-source repository that can be confirmed to belong to OhYesAI has been found at present.
The integration of third-party generation models by a platform does not mean that its own source code, workflows, or services are open source.
Usage restrictions
- Creating a long MV requires more points and a longer waiting time.
- The characters’ faces, clothing, and movements may change from one shot to another.
- Automatic timing may still require manual adjustment.
- Text generation, musical instrument performance, and complex movements may be distorted.
- The expected output is likely to vary depending on the model, duration, and number of retries.
- A commercial license does not eliminate the risks associated with using imported music and reference images.
Frequently Asked Questions
Can OhYesAI be used for free?
Sure, registration comes with 1500 points as a gift, allowing you to experience a MV with watermarks for about 15 to 30 seconds.
How long can a music MV be for OhYesAI?
The paid version allows for a full 5-minute video, while the free trial version provides a duration of around 15 to 30 seconds.
Can I upload my own songs?
You can upload local MP3 files, or use the platform’s built-in AI music feature to create original songs.
Can I control the characters and visual style?
Yes, the official version supports one to seven reference images, which can be used to determine the character, clothing, environment, and props.
How much is the basic version of OhYesAI?
The official website’s current monthly fee is 78 yuan; the annual fee is equivalent to 65 yuan per month, and it includes 7,800 points per month.
Can the generated MV be used for commercial purposes?
The paid package includes watermark removal and commercial licensing; the content generated through the free trial cannot be used for commercial purposes.
Is OhYesAI open source?
It is not open source; at present, no official complete source code repository or open deployment version has been identified.
Guigong Network Security Registration No. 45132202000164