Piano Music Big Model - AI Music Composition Model by Tencent

What is the Zinle Grand Model

Qin Music Grand Model is an advanced AI music creation grand model jointly launched by Tencent AI Lab and Tencent TME Tianqin Lab. The model intelligently generates high-quality stereo audio or multi-track sheet music based on user input of Chinese and English keywords, descriptive statements or audio clips. The model supports automatic editing functions, such as renewing the score, regenerating specified tracks or measures, adjusting the orchestration, modifying instrument types and rhythms, and ensuring the relevance and accuracy of the generated music based on the audio text alignment technology and the prediction of the big language model. The Qin Music Big Model follows music theory to ensure that elements such as melody, chords, and beats are in line with human aesthetics. The model is now online on the Tencent Music Qixing platform, where users can visit to experience the model's powerful music creation capabilities.

Key Features of the Zither Grand Model

• Music Generation: Generate high-quality stereo audio based on keywords, descriptive statements or audio input in English and Chinese.
• Score Generation: Generate detailed scores with multiple tracks of melody, chords, accompaniment and percussion.
• automatic editingSupports continuation, regeneration of a given track or measure, and modification of instrument types and tempo.
• Audio Text Alignment: Understanding the semantic relationships between text descriptions and audio ensures the relevance and accuracy of the generated music.
• Score/Audio Characterization Extraction: Converting musical scores or audio into discrete feature sequences to provide a basis for large language model predictions.
• Large Language Model Prediction: Intelligent conversion from text to music based on feature prediction training.
• Audio Recovery: Recovering predicted audio representation sequences into high-quality audible audio using stream matching and vocoder techniques.
• Music theory follows: The music generated follows the music theory of melody, chords, and beats and conforms to human aesthetics.

The official website of Ginra Grand Models

• Official website address::Tencent Music - Qixing

How to use the Zither Big Model

• Access platforms：Login to Tencent Music - Qixing Platform'sOfficial websiteThe
• Register/Login: Create a new account or log in with an existing account.
• Enter creation information: Enter musical keywords, descriptive statements or upload audio clips on the page as a basis for creation.
• Select Model & Duration: Select Piano Music Grand Model v1.0 and set the music duration (10 seconds to 30 seconds).
• Generate MusicClick the "Generate" button, AI automatically generates the music composition, support playback and download.

Core Strengths of Ginra's Big Models

• Efficient creation: Quickly generate music and support automatic editing to save time.
• High quality output: Generate music that is highly relevant to the input, follows music theory, and conforms to human aesthetics.
• Flexible customization: Supports multiple input methods to generate multi-track scores to meet personalized needs.
• technologically advanced: Ensure musical coherence and realism with macro-language modeling predictions and audio restoration techniques.
• user experience: End-to-end generation process, integrated authoring experience, validated by large-scale double-blind listening tests.

People for whom the Ginra Grand Model is suitable

• professional musician: Generate music and scores quickly, providing creative inspiration and saving time.
• hobbyist: Easy to create personalized music without specialized skills.
• content creator: Create background music for videos, commercials, etc. to enhance the quality of content.
• music educator: Used in teaching to help students understand music theory and the creative process.