InstructME: An Instruction Guided Music Edit Framework with Latent Diffusion Models

Abstract

Music editing primarily entails the modification of instrument tracks or remixing in the whole, which offers a novel reinterpretation of the original piece through a series of operations. These music processing methods hold immense potential across various applications but demand substantial expertise. Prior methodologies, although effective for image and audio modifications, falter when directly applied to music. This is attributed to music's distinctive data nature, where such methods can inadvertently compromise the intrinsic harmony and coherence of music. In this paper, we develop InstructME, an Instruction guided Music Editing and remixing framework based on latent diffusion models. Our framework fortifies the U-Net with multi-scale aggregation in order to maintain consistency before and after editing. In addition, we introduce chord progression matrix as condition information and incorporate it in the semantic space to improve melodic harmony while editing. For accommodating extended musical pieces, InstructME employs a chunk transformer, enabling it to discern long-term temporal dependencies within music sequences. We tested InstructME in instrument-editing, remixing, and multi-round editing. Both subjective and objective evaluations indicate that our proposed method significantly surpasses preceding systems in music quality, text relevance and harmony.

Overview of InstructME

Left: Overview of InstructME diffusion process for music editing. Audio signal is processed by VAE (encoder `\mathcal{E}` and decoder `\mathcal{D}` ), meanwhile extractor `\mathcal{C}` extracts the chord matrix of source music and together with text embedding extracted by `\mathcal{T}` as condition information, latent embedding `z_{s}` and `z_{t}` are fused by multi-scale aggregation and converted by chunk transformer to produce the final edited music. Right: Architecture of chunk transformer(C-T) blocks which in various positions of U-net will selectively incorporate chord or text embedding, and `z_{s}` will only input when chunk transformer is in down sampler.

Video introduction of InstructME

Samples for atomic editing operations

Our InstructME supports atomic editing operations on music, including adding, removing, extracting, and replacing instruments.
Text Prompt: The command used for editing.
Source: The music before editing.
Target: The groundtruth after music editing.
InstructME and AUDIT: The edited music generated by our proposed InstructME and baseline AUDIT respectively.

Atomic operations - Add

Text Prompt Source InstructME AUDIT Target

Atomic operations - Remove

Text Prompt Source InstructME AUDIT Target

Atomic operations - Extract

Text Prompt Source InstructME AUDIT Target

Atomic operations - Replace

Text Prompt Source InstructME AUDIT Target

Samples for Remix operations

Remixing can be understood as an advanced version of music editing that mixes various atomic operations with style and genre considered.
accompliment is accompany only and original music is accompany with vocal. The harmony of vocals can better evaluate the chord consistency of edited music.

Remix

Text Prompt Source-accompliment Source-original music InstructME-accompliment InstructME-original music AUDIT-accompliment AUDIT-original music Target-accompliment Target-original music

Remix with genres

Text Prompt Source-accompliment Source-original music InstructME-accompliment InstructME-original music AUDIT-accompliment AUDIT-original music Target-accompliment Target-original music

Remix and Guided to soft mood music

Use guidance to control the mood of edited music. Guidance prompt: "a soft music".
Text Prompt Source-accompliment Source-original music InstructME-accompliment InstructME-original music AUDIT-accompliment AUDIT-original music Target-accompliment Target-original music

Remix and Guided to happy mood music

Use guidance to control the mood of edited music. Guidance prompt: "a happy music".
Text Prompt Source-accompliment Source-original music InstructME-accompliment InstructME-original music AUDIT-accompliment AUDIT-original music Target-accompliment Target-original music

Diversity and Stability

Different editing operations require different modeling capabilities.
For creativity-oriented tasks including remixing, adding and replacing, our proposed InstructME can generate diverse edited results.
For tasks requiring precision including extracting and removing, our proposed InstructME can consistently generates results congruent with the ground truth.

Diversity in atomic editing tasks

Text Prompt Source InstructME-1 InstructME-2 InstructME-3 Target

Diversity in remix tasks

Text Prompt Source-accompliment Source-original music InstructME-accompliment-1 InstructME-original music-1 InstructME-accompliment-2 InstructME-original music-2 InstructME-accompliment-3 InstructME-original music-3 Target-accompliment Target-original music

Stability in atomic editing tasks

Text Prompt Source InstructME-1 InstructME-2 InstructME-3 Target

Real Data

We provide some examples with editing real music data.

Real song editing

Text Prompt Song Title Source InstructME

Real song editing with vocal

Text Prompt Song Title Source InstructME

Multi-round Editing

Due to the consistency and harmony of our editing model, it also supports multiple rounds of editing.
Method Source Command 1: Add acoustic guitar Command 2: Add drum kit Command 3: replace acoustic guitar with piano

Long Music Editing

InstructME supports long music editing.
Text Prompt Source Duration InstructME AUDIT