Tempolor v4.7 - A flagship AI music generation model launched by Quwan Technology.

What is Tempolor v4.7?

Tempolor v4.7 is a flagship AI music generation model launched by Quwan Technology. It adopts a fourth-generation layered progressive architecture and supports 48kHz dual-channel high-quality output. The model focuses on controllable generation and secondary creation, enabling style rewriting, cover songs, and fine-grained audio editing while retaining the core melody. It goes from one-click song creation to free modification, and is open to developers and ordinary users through OpenAPI and the Tunee intelligent agent.

Main features of Tempolor v4.7

• Remix rewriteUpload existing music materials, retain the core melody and motif, and redesign the style, arrangement, and singing method.
• Cover SongThe lyrics and vocal timbre were changed, and the original song's recognizability was maintained through the "rapid return motif" technique to avoid melody misalignment.
• Fine-grained audio editing: Precisely adjust instrument timbre, harmony, vocals, and backing vocals using prompts.
• Reference Audio Style TransferUpload reference audio to generate new songs in the same style, while optimizing audio leakage and outputting a more independent and clean sound.
• Multimodal generationIt supports generating music using various input methods such as text, images, and videos.
• Dialogue-based creation: Connect to the Tunee AI agent to complete music creation and visual generation through dialogue.

Technical Principles of Tempolor v4.7

• Fourth-generation layered progressive architectureThe music generation process is broken down into three levels: musicality, semantics, and acoustics. These are completed sequentially from coarse to fine. First, the theme, motif, and overall structure are determined. Then, the specific content is generated, and finally, vocal and instrumental details are added.
• Rapid regression motivationIn cover song scenarios, the model appropriately rewrites the lyrics based on the new lyric structure, while refocusing on the main melody at key points, balancing creative freedom with the original song's recognizability.
• Reference audio leak optimizationWhen the model learns the original melody structure, it reduces the direct copying of the original vocals and timbre.
• Tag system reconstruction: Readjust the training data labels to improve the accuracy of recognition and response to information such as instrument timbre, harmony, and vocals.
• Continuous acoustic characterizationA continuous audio representation scheme combining masked autoregression and diffusion models is adopted to balance generation quality and local modification capability.

How to use Tempolor v4.7

• Regular usersAccess the Tunee agent and input your creative needs through dialogue or by uploading materials.
• Developers/EnterprisesAccess the Tianpule OpenAPI platform and call the V4.7 capabilities through the interface to connect your own applications or workflows.
• CreatorsUpload reference audio or enter prompt words, select Remix, Cover, or a new generation mode, and continuously modify and iterate the output.

The core advantages of Tempolor v4.7

• Highly controllableThe hierarchical architecture makes the prompt word response more stable, the output results of the same input are more deterministic, and the cost of repeated generation and filtering is reduced.
• Leading in secondary creationIt boasts leading domestic capabilities in remixing and cover songs, allowing for free adjustments while retaining the core motif.
• Excellent sound quality48kHz dual-channel output, improved spatial separation of drums, bass, harmonies and vocals, and more delicate emotional expression in slow-tempo music.
• Multimodal inputThe world's first large-scale music generation model that supports multimodal input of text, images, audio, and video.
• Performance improvementOverall performance improved by about 20%, and the first packet response time for streaming generation was about 20 seconds.

Comparison of Tempolor v4.7 with similar competing products

Application scenarios of Tempolor v4.7

• Short video background musicCreators can quickly generate or adapt background music to match the mood of their videos, and can make multiple fine-tuning adjustments until they are satisfied.
• Advertising and FilmGenerate customized music for brand advertisements and film clips, and quickly adapt to different version requirements through Remix.
• Music EducationStudents upload their practice pieces, and AI assists in adapting them into different styles, lowering the learning threshold for music arrangement.
• Independent musician creationMusicians use AI-generated results as creative material, continuously modifying and processing them to create complete works.
• Game audioGenerate theme music for game scenes and quickly adapt theme songs for multiple languages using the Cover feature.