paper-with-me

홈 › Papers

ETTA: Elucidating the Design Space of Text-to-Audio Models

2024-12-26 · Sang-gil Lee, Zhifeng Kong, Arushi Goel, Sungwon Kim, Rafael Valle, Bryan Catanzaro

Recent years have seen significant progress in Text-To-Audio (TTA) synthesis, enabling users to enrich their creative workflows with synthetic audio generated from natural language prompts. Despite this progress, the effects of data, model architecture, training objective functions, and sampling strategies on target benchmarks are not well understood. With the purpose of providing a holistic understanding of the design space of TTA models, we set up a large-scale empirical experiment focused on diffusion and flow matching models. Our contributions include: 1) AF-Synthetic, a large dataset of high quality synthetic captions obtained from an audio understanding model; 2) a systematic comparison of different architectural, training, and inference design choices for TTA models; 3) an analysis of sampling methods and their Pareto curves with respect to generation quality and inference speed. We leverage the knowledge obtained from this extensive analysis to propose our best model dubbed Elucidated Text-To-Audio (ETTA). When evaluated on AudioCaps and MusicCaps, ETTA provides improvements over the baselines trained on publicly available data, while being competitive with models trained on proprietary data. Finally, we show ETTA's improved ability to generate creative audio following complex and imaginative captions -- a task that is more challenging than current benchmarks.

📄 PDF Abstract BibTeX arXiv:2412.19351

Code (1)

NVIDIA/elucidated-text-to-audio

Tasks

AudioCapsAudio captioningAudio GenerationLanguage ModellingMusic GenerationText-to-Music Generation

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

VoiceTTA: Enhancing Zero-Shot Text-to-Speech via Reinforcement Learning-Based Test-Time Adaptation

2026-06-25 · Tianxin Xie, Chenxing Li, Dong Yu, Li Liu arxiv

Recently, zero-shot text-to-speech (TTS) has enabled high-fidelity and expressive speech synthesis, but it often fails to imitate unseen speaking styles from uncommon scenarios (e.g., crosstalk, dialects). Moreover, fine…

Reinforcement LearningTest-time AdaptationSpeech Synthesis

Training Transitive and Commutative Multimodal Transformers with LoReTTa

2023-05-23 · NeurIPS 2023 11 · Manuel Tran, Yashin Dicente Cid, Amal Lahiani, Fabian J. Theis 외

Training multimodal foundation models is challenging due to the limited availability of multimodal datasets. While many public datasets pair images with text, few combine images with audio or text with audio. Even rarer …

Triplet

RosettaSearch: Multi-Objective Inference-Time Search for Protein Sequence Design

2026-04-19 · Meghana Kshirsagar, Allen Nie, Ching-An Cheng, Fanglei Xue 외 arxiv

We introduce RosettaSearch, an inference-time multi-objective optimization approach for backbone conditioned protein sequence design. We use large language models (LLMs) as a generative optimizer within a search algorith…

Protein Design with Agent Rosetta: A Case Study for Specialized Scientific Agents

2026-03-16 · Jacopo Teneggi, S. M. Bargeen A. Turzo, Tanya Marwah, Alberto Bietti 외 arxiv

Large language models (LLMs) are capable of emulating reasoning and using tools, creating opportunities for autonomous agents that execute complex scientific tasks. Protein design provides a natural testbed: although mac…

Prompt EngineeringProtein Design

NetTAG: A Multimodal RTL-and-Layout-Aligned Netlist Foundation Model via Text-Attributed Graph

2025-04-12 · Wenji Fang, Wenkai Li, Shang Liu, Yao Lu 외

Circuit representation learning has shown promise in advancing Electronic Design Automation (EDA) by capturing structural and functional circuit properties for various tasks. Existing pre-trained solutions rely on graph …

Graph LearningRepresentation Learning