Open-source music generation model series, end-to-end music generation, open-source text-to-music generation series

Open-Source AI Music Generation Models

1. DiffRhythm

• Description: An end-to-end music model based on diffusion models, capable of generating a complete song with vocals and accompaniment within 10 seconds.
• Demo: Details and demonstration

2. InspireMusic

• Description: An open-source music generation model by Alibaba, allowing control over text prompts, music genres, and music structure.
• Demo: Details and demonstration

3. YuE

• Description: An open-source music foundation model from the University of Hong Kong, based on the LLaMA architecture, capable of generating songs up to 5 minutes long.
• Demo: Details and demonstration

4. MelodyFlow

• Description: An AI music generation model released by Meta, capable of generating music based on prompt words and supporting online music editing.
• Demo: Details and demonstration

5. QA-MDT

• Description: An open-source text-to-music generation model based on diffusion models.
• Demo: Details and demonstration

6. Seed-Music

• Description: A unified AI music generation framework by DouBao, more comprehensive and advanced than Suno.
• Demo: Details and demonstration

7. FluxMusic

• Description: A text-to-music generation model based on Transformer.
• Demo: Details and demonstration

8. MusiConGen

• Description: A text-to-music generation model based on Transformer, allowing control over chords and tempo (BPM) through text.
• Demo: Details and demonstration

9. EzAudio

• Description: An open-source AI audio generation model, enhancing text-to-audio generation based on Diffusion Transformer.
• Demo: Details and demonstration