Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models

Training-free guidance enables controlled generation in diffusion and flow models, but most existing methods assume differentiable objectives and rely on gradients. This work focuses on training-free guidance addressing challenges from non-differentiable objectives and discrete data distributions. We propose an algorithmic framework TreeG: Tree Search-Based Path Steering Guidance, applicable to both continuous and discrete settings in diffusion and flow models. TreeG offers a unified perspective on training-free guidance: proposing candidates for the next step, evaluating candidates, and selecting the best to move forward, enhanced by a tree search mechanism over active paths or parallelizing exploration. We comprehensively investigate the design space of TreeG over the candidate proposal module and the evaluation function, instantiating TreeG into three novel algorithms. Our experiments show that TreeG consistently outperforms the top guidance baselines in symbolic music generation, small molecule generation, and enhancer DNA design, all of which involve non-differentiable challenges. Additionally, we identify an inference-time scaling law showing TreeG's scalability in inference-time computation.

Global Position Aware Group Choreography using Large Language Model

Dance serves as a profound and universal expression of human culture, conveying emotions and stories through movements synchronized with music. Although some current works have achieved satisfactory results in the task of single-person dance generation, the field of multi-person dance generation remains relatively novel. In this work, we present a group choreography framework that leverages recent advancements in Large Language Models (LLM) by modeling the group dance generation problem as a sequence-to-sequence translation task. Our framework consists of a tokenizer that transforms continuous features into discrete tokens, and an LLM that is fine-tuned to predict motion tokens given the audio tokens. We show that by proper tokenization of input modalities and careful design of the LLM training strategies, our framework can generate realistic and diverse group dances while maintaining strong music correlation and dancer-wise consistency. Extensive experiments and evaluations demonstrate that our framework achieves state-of-the-art performance.

TokenSynth: A Token-based Neural Synthesizer for Instrument Cloning and Text-to-Instrument

Recent advancements in neural audio codecs have enabled the use of tokenized audio representations in various audio generation tasks, such as text-to-speech, text-to-audio, and text-to-music generation. Leveraging this approach, we propose TokenSynth, a novel neural synthesizer that utilizes a decoder-only transformer to generate desired audio tokens from MIDI tokens and CLAP (Contrastive Language-Audio Pretraining) embedding, which has timbre-related information. Our model is capable of performing instrument cloning, text-to-instrument synthesis, and text-guided timbre manipulation without any fine-tuning. This flexibility enables diverse sound design and intuitive timbre control. We evaluated the quality of the synthesized audio, the timbral similarity between synthesized and target audio/text, and synthesis accuracy (i.e., how accurately it follows the input MIDI) using objective measures. TokenSynth demonstrates the potential of leveraging advanced neural audio codecs and transformers to create powerful and versatile neural synthesizers. The source code, model weights, and audio demos are available at: https://github.com/KyungsuKim42/tokensynth

Designing Neural Synthesizers for Low Latency Interaction

Neural Audio Synthesis (NAS) models offer interactive musical control over high-quality, expressive audio generators. While these models can operate in real-time, they often suffer from high latency, making them unsuitable for intimate musical interaction. The impact of architectural choices in deep learning models on audio latency remains largely unexplored in the NAS literature. In this work, we investigate the sources of latency and jitter typically found in interactive NAS models. We then apply this analysis to the task of timbre transfer using RAVE, a convolutional variational autoencoder for audio waveforms introduced by Caillon et al. in 2021. Finally, we present an iterative design approach for optimizing latency. This culminates with a model we call BRAVE (Bravely Realtime Audio Variational autoEncoder), which is low-latency and exhibits better pitch and loudness replication while showing timbre modification capabilities similar to RAVE. We implement it in a specialized inference framework for low-latency, real-time inference and present a proof-of-concept audio plugin compatible with audio signals from musical instruments. We expect the challenges and guidelines described in this document to support NAS researchers in designing models for low-latency inference from the ground up, enriching the landscape of possibilities for musicians.

TALKPLAY: Multimodal Music Recommendation with Large Language Models

We present TalkPlay, a multimodal music recommendation system that reformulates the recommendation task as large language model token generation. TalkPlay represents music through an expanded token vocabulary that encodes multiple modalities - audio, lyrics, metadata, semantic tags, and playlist co-occurrence. Using these rich representations, the model learns to generate recommendations through next-token prediction on music recommendation conversations, that requires learning the associations natural language query and response, as well as music items. In other words, the formulation transforms music recommendation into a natural language understanding task, where the model's ability to predict conversation tokens directly optimizes query-item relevance. Our approach eliminates traditional recommendation-dialogue pipeline complexity, enabling end-to-end learning of query-aware music recommendations. In the experiment, TalkPlay is successfully trained and outperforms baseline methods in various aspects, demonstrating strong context understanding as a conversational music recommender.

Audio-FLAN: A Preliminary Release

Recent advancements in audio tokenization have significantly enhanced the integration of audio capabilities into large language models (LLMs). However, audio understanding and generation are often treated as distinct tasks, hindering the development of truly unified audio-language models. While instruction tuning has demonstrated remarkable success in improving generalization and zero-shot learning across text and vision, its application to audio remains largely unexplored. A major obstacle is the lack of comprehensive datasets that unify audio understanding and generation. To address this, we introduce Audio-FLAN, a large-scale instruction-tuning dataset covering 80 diverse tasks across speech, music, and sound domains, with over 100 million instances. Audio-FLAN lays the foundation for unified audio-language models that can seamlessly handle both understanding (e.g., transcription, comprehension) and generation (e.g., speech, music, sound) tasks across a wide range of audio domains in a zero-shot manner. The Audio-FLAN dataset is available on HuggingFace and GitHub and will be continuously updated.

A Reservoir-based Model for Human-like Perception of Complex Rhythm Pattern

Rhythm is a fundamental aspect of human behaviour, present from infancy and deeply embedded in cultural practices. Rhythm anticipation is a spontaneous cognitive process that typically occurs before the onset of actual beats. While most research in both neuroscience and artificial intelligence has focused on metronome-based rhythm tasks, studies investigating the perception of complex musical rhythm patterns remain limited. To address this gap, we propose a hierarchical oscillator-based model to better understand the perception of complex musical rhythms in biological systems. The model consists of two types of coupled neurons that generate oscillations, with different layers tuned to respond to distinct perception levels. We evaluate the model using several representative rhythm patterns spanning the upper, middle, and lower bounds of human musical perception. Our findings demonstrate that, while maintaining a high degree of synchronization accuracy, the model exhibits human-like rhythmic behaviours. Additionally, the beta band neuronal activity in the model mirrors patterns observed in the human brain, further validating the biological plausibility of the approach.

Exploring the Needs of Practising Musicians in Co-Creative AI Through Co-Design

Recent advances in generative AI music have resulted in new technologies that are being framed as co-creative tools for musicians with early work demonstrating their potential to add to music practice. While the field has seen many valuable contributions, work that involves practising musicians in the design and development of these tools is limited, with the majority of work including them only once a tool has been developed. In this paper, we present a case study that explores the needs of practising musicians through the co-design of a musical variation system, highlighting the importance of involving a diverse range of musicians throughout the design process and uncovering various design insights. This was achieved through two workshops and a two week ecological evaluation, where musicians from different musical backgrounds offered valuable insights not only on a musical system's design but also on how a musical AI could be integrated into their musical practices.

Réduire le bruit grâce à la réalité augmentée sonore -- Auditory Concealer

This report presents the work done over 22 weeks of internship within the Sound Perception and Design team of the Sciences and Technologies of Music and Sound (STMS) laboratory at the Institute for Research and Coordination in Acoustics/Music (IRCAM). As part of the launch of the project Reducing Noise with Augmented Reality (ReNAR); which aims to create a tool to reduce in real-time the cognitive impact of sounds perceived as unpleasant or annoying in indoor environments; an initial study was conducted to validate the feasibility and effectiveness of a new masking approach called concealer. The main hypothesis is that the concealer approach could provide better results than a masker approach in terms of perceived pleasantness. Mixtures of two noise sources (ventilation) and five masking sounds (water sounds) were generated using both approaches at various levels. The evaluation of the perceived pleasantness of these mixtures showed that the masker approach remains more effective than the concealer approach, regardless of the noise source, water sound, or level used.

Unsupervised Estimation of Nonlinear Audio Effects: Comparing Diffusion-Based and Adversarial approaches

Accurately estimating nonlinear audio effects without access to paired input-output signals remains a challenging problem.This work studies unsupervised probabilistic approaches for solving this task. We introduce a method, novel for this application, based on diffusion generative models for blind system identification, enabling the estimation of unknown nonlinear effects using black- and gray-box models. This study compares this method with a previously proposed adversarial approach, analyzing the performance of both methods under different parameterizations of the effect operator and varying lengths of available effected recordings.Through experiments on guitar distortion effects, we show that the diffusion-based approach provides more stable results and is less sensitive to data availability, while the adversarial approach is superior at estimating more pronounced distortion effects. Our findings contribute to the robust unsupervised blind estimation of audio effects, demonstrating the potential of diffusion models for system identification in music technology.