We introduce MDSGen, a novel framework for vision-guided open-domain sound generation optimized for model parameter size, memory consumption, and inference speed. This framework incorporates two key innovations: (1) a redundant video feature removal module that filters out unnecessary visual information, and (2) a temporal-aware masking strategy that leverages temporal context for enhanced audio generation accuracy. In contrast to existing resource-heavy Unet-based models, \texttt{MDSGen} employs denoising masked diffusion transformers, facilitating efficient generation without reliance on pre-trained diffusion models. Evaluated on the benchmark VGGSound dataset, our smallest model (5M parameters) achieves $97.9$% alignment accuracy, using $172\times$ fewer parameters, $371$% less memory, and offering $36\times$ faster inference than the current 860M-parameter state-of-the-art model ($93.9$% accuracy). The larger model (131M parameters) reaches nearly $99$% accuracy while requiring $6.5\times$ fewer parameters. These results highlight the scalability and effectiveness of our approach. The code is available at https://bit.ly/mdsgen.
Building upon advancements in Large Language Models (LLMs), the field of audio processing has seen increased interest in training audio generation tasks with discrete audio token sequences. However, directly discretizing audio by neural audio codecs often results in sequences that fundamentally differ from text sequences. Unlike text, where text token sequences are deterministic, discrete audio tokens can exhibit significant variability based on contextual factors, while still producing perceptually identical audio segments. We refer to this phenomenon as \textbf{Discrete Representation Inconsistency (DRI)}. This inconsistency can lead to a single audio segment being represented by multiple divergent sequences, which creates confusion in neural codec language models and results in omissions and repetitions during speech generation. In this paper, we quantitatively analyze the DRI phenomenon within popular audio tokenizers such as EnCodec. Our approach effectively mitigates the DRI phenomenon of the neural audio codec. Furthermore, extensive experiments on the neural codec language model over LibriTTS and large-scale MLS datases (44,000 hours) demonstrate the effectiveness and generality of our method. The demo of audio samples is available online~\footnote{\url{https://consistencyinneuralcodec.github.io}}.
Neural codecs have become crucial to recent speech and audio generation research. In addition to signal compression capabilities, discrete codecs have also been found to enhance downstream training efficiency and compatibility with autoregressive language models. However, as extensive downstream applications are investigated, challenges have arisen in ensuring fair comparisons across diverse applications. To address these issues, we present a new open-source platform ESPnet-Codec, which is built on ESPnet and focuses on neural codec training and evaluation. ESPnet-Codec offers various recipes in audio, music, and speech for training and evaluation using several widely adopted codec models. Together with ESPnet-Codec, we present VERSA, a standalone evaluation toolkit, which provides a comprehensive evaluation of codec performance over 20 audio evaluation metrics. Notably, we demonstrate that ESPnet-Codec can be integrated into six ESPnet tasks, supporting diverse applications.
Built upon vector quantization (VQ), discrete audio codec models have achieved great success in audio compression and auto-regressive audio generation. However, existing models face substantial challenges in perceptual quality and signal distortion, especially when operating in extremely low bandwidth, rooted in the sensitivity of the VQ codebook to noise. This degradation poses significant challenges for several downstream tasks, such as codec-based speech synthesis. To address this issue, we propose a novel VQ method, Normal Distribution-based Vector Quantization (NDVQ), by introducing an explicit margin between the VQ codes via learning a variance. Specifically, our approach involves mapping the waveform to a latent space and quantizing it by selecting the most likely normal distribution, with each codebook entry representing a unique normal distribution defined by its mean and variance. Using these distribution-based VQ codec codes, a decoder reconstructs the input waveform. NDVQ is trained with additional distribution-related losses, alongside reconstruction and discrimination losses. Experiments demonstrate that NDVQ outperforms existing audio compression baselines, such as EnCodec, in terms of audio quality and zero-shot TTS, particularly in very low bandwidth scenarios.
Current Text-to-audio (TTA) models mainly use coarse text descriptions as inputs to generate audio, which hinders models from generating audio with fine-grained control of content and style. Some studies try to improve the granularity by incorporating additional frame-level conditions or control networks. However, this usually leads to complex system design and difficulties due to the requirement for reference frame-level conditions. To address these challenges, we propose AudioComposer, a novel TTA generation framework that relies solely on natural language descriptions (NLDs) to provide both content specification and style control information. To further enhance audio generative modeling, we employ flow-based diffusion transformers with the cross-attention mechanism to incorporate text descriptions effectively into audio generation processes, which can not only simultaneously consider the content and style information in the text inputs, but also accelerate generation compared to other architectures. Furthermore, we propose a novel and comprehensive automatic data simulation pipeline to construct data with fine-grained text descriptions, which significantly alleviates the problem of data scarcity in the area. Experiments demonstrate the effectiveness of our framework using solely NLDs as inputs for content specification and style control. The generation quality and controllability surpass state-of-the-art TTA models, even with a smaller model size.
Video encompasses both visual and auditory data, creating a perceptually rich experience where these two modalities complement each other. As such, videos are a valuable type of media for the investigation of the interplay between audio and visual elements. Previous studies of audio-visual modalities primarily focused on either audio-visual representation learning or generative modeling of a modality conditioned on the other, creating a disconnect between these two branches. A unified framework that learns representation and generates modalities has not been developed yet. In this work, we introduce a novel framework called Vision to Audio and Beyond (VAB) to bridge the gap between audio-visual representation learning and vision-to-audio generation. The key approach of VAB is that rather than working with raw video frames and audio data, VAB performs representation learning and generative modeling within latent spaces. In particular, VAB uses a pre-trained audio tokenizer and an image encoder to obtain audio tokens and visual features, respectively. It then performs the pre-training task of visual-conditioned masked audio token prediction. This training strategy enables the model to engage in contextual learning and simultaneous video-to-audio generation. After the pre-training phase, VAB employs the iterative-decoding approach to rapidly generate audio tokens conditioned on visual features. Since VAB is a unified model, its backbone can be fine-tuned for various audio-visual downstream tasks. Our experiments showcase the efficiency of VAB in producing high-quality audio from video, and its capability to acquire semantic audio-visual features, leading to competitive results in audio-visual retrieval and classification.
We introduce V-AURA, the first autoregressive model to achieve high temporal alignment and relevance in video-to-audio generation. V-AURA uses a high-framerate visual feature extractor and a cross-modal audio-visual feature fusion strategy to capture fine-grained visual motion events and ensure precise temporal alignment. Additionally, we propose VisualSound, a benchmark dataset with high audio-visual relevance. VisualSound is based on VGGSound, a video dataset consisting of in-the-wild samples extracted from YouTube. During the curation, we remove samples where auditory events are not aligned with the visual ones. V-AURA outperforms current state-of-the-art models in temporal alignment and semantic relevance while maintaining comparable audio quality. Code, samples, VisualSound and models are available at https://v-aura.notion.site
With recent advances of AIGC, video generation have gained a surge of research interest in both academia and industry (e.g., Sora). However, it remains a challenge to produce temporally aligned audio to synchronize the generated video, considering the complicated semantic information included in the latter. In this work, inspired by the recent success of text-to-audio (TTA) generation, we first investigate the video-to-audio (VTA) generation framework based on latent diffusion model (LDM). Similar to latest pioneering exploration in VTA, our preliminary results also show great potentials of LDM in VTA task, but it still suffers from sub-optimal temporal alignment. To this end, we propose to enhance the temporal alignment of VTA with frame-level semantic information. With the recently popular grounding segment anything model (Grounding SAM), we can extract the fine-grained semantics in video frames to enable VTA to produce better-aligned audio signal. Extensive experiments demonstrate the effectiveness of our system on both objective and subjective evaluation metrics, which shows both better audio quality and fine-grained temporal alignment.
Denoising diffusion models have emerged as state-of-the-art in generative tasks across image, audio, and video domains, producing high-quality, diverse, and contextually relevant data. However, their broader adoption is limited by high computational costs and large memory footprints. Post-training quantization (PTQ) offers a promising approach to mitigate these challenges by reducing model complexity through low-bandwidth parameters. Yet, direct application of PTQ to diffusion models can degrade synthesis quality due to accumulated quantization noise across multiple denoising steps, particularly in conditional tasks like text-to-audio synthesis. This work introduces PTQ4ADM, a novel framework for quantizing audio diffusion models(ADMs). Our key contributions include (1) a coverage-driven prompt augmentation method and (2) an activation-aware calibration set generation algorithm for text-conditional ADMs. These techniques ensure comprehensive coverage of audio aspects and modalities while preserving synthesis fidelity. We validate our approach on TANGO, Make-An-Audio, and AudioLDM models for text-conditional audio generation. Extensive experiments demonstrate PTQ4ADM's capability to reduce the model size by up to 70\% while achieving synthesis quality metrics comparable to full-precision models($<$5\% increase in FD scores). We show that specific layers in the backbone network can be quantized to 4-bit weights and 8-bit activations without significant quality loss. This work paves the way for more efficient deployment of ADMs in resource-constrained environments.
We introduce EzAudio, a text-to-audio (T2A) generation framework designed to produce high-quality, natural-sounding sound effects. Core designs include: (1) We propose EzAudio-DiT, an optimized Diffusion Transformer (DiT) designed for audio latent representations, improving convergence speed, as well as parameter and memory efficiency. (2) We apply a classifier-free guidance (CFG) rescaling technique to mitigate fidelity loss at higher CFG scores and enhancing prompt adherence without compromising audio quality. (3) We propose a synthetic caption generation strategy leveraging recent advances in audio understanding and LLMs to enhance T2A pretraining. We show that EzAudio, with its computationally efficient architecture and fast convergence, is a competitive open-source model that excels in both objective and subjective evaluations by delivering highly realistic listening experiences. Code, data, and pre-trained models are released at: https://haidog-yaqub.github.io/EzAudio-Page/.
Neural audio codecs have significantly advanced audio compression by efficiently converting continuous audio signals into discrete tokens. These codecs preserve high-quality sound and enable sophisticated sound generation through generative models trained on these tokens. However, existing neural codec models are typically trained on large, undifferentiated audio datasets, neglecting the essential discrepancies between sound domains like speech, music, and environmental sound effects. This oversight complicates data modeling and poses additional challenges to the controllability of sound generation. To tackle these issues, we introduce the Source-Disentangled Neural Audio Codec (SD-Codec), a novel approach that combines audio coding and source separation. By jointly learning audio resynthesis and separation, SD-Codec explicitly assigns audio signals from different domains to distinct codebooks, sets of discrete representations. Experimental results indicate that SD-Codec not only maintains competitive resynthesis quality but also, supported by the separation results, demonstrates successful disentanglement of different sources in the latent space, thereby enhancing interpretability in audio codec and providing potential finer control over the audio generation process.
Current mainstream audio generation methods primarily rely on simple text prompts, often failing to capture the nuanced details necessary for multi-style audio generation. To address this limitation, the Sound Event Enhanced Prompt Adapter is proposed. Unlike traditional static global style transfer, this method extracts style embedding through cross-attention between text and reference audio for adaptive style control. Adaptive layer normalization is then utilized to enhance the model's capacity to express multiple styles. Additionally, the Sound Event Reference Style Transfer Dataset (SERST) is introduced for the proposed target style audio generation task, enabling dual-prompt audio generation using both text and audio references. Experimental results demonstrate the robustness of the model, achieving state-of-the-art Fr\'echet Distance of 26.94 and KL Divergence of 1.82, surpassing Tango, AudioLDM, and AudioGen. Furthermore, the generated audio shows high similarity to its corresponding audio reference. The demo, code, and dataset are publicly available.
We propose TSELM, a novel target speaker extraction network that leverages discrete tokens and language models. TSELM utilizes multiple discretized layers from WavLM as input tokens and incorporates cross-attention mechanisms to integrate target speaker information. Language models are employed to capture the sequence dependencies, while a scalable HiFi-GAN is used to reconstruct the audio from the tokens. By applying a cross-entropy loss, TSELM models the probability distribution of output tokens, thus converting the complex regression problem of audio generation into a classification task. Experimental results show that TSELM achieves excellent results in speech quality and comparable results in speech intelligibility.
This paper introduces MetaBGM, a groundbreaking framework for generating background music that adapts to dynamic scenes and real-time user interactions. We define multi-scene as variations in environmental contexts, such as transitions in game settings or movie scenes. To tackle the challenge of converting backend data into music description texts for audio generation models, MetaBGM employs a novel two-stage generation approach that transforms continuous scene and user state data into these texts, which are then fed into an audio generation model for real-time soundtrack creation. Experimental results demonstrate that MetaBGM effectively generates contextually relevant and dynamic background music for interactive applications.
In recent years, artificial intelligence (AI) has made significant progress in the field of music generation, driving innovation in music creation and applications. This paper provides a systematic review of the latest research advancements in AI music generation, covering key technologies, models, datasets, evaluation methods, and their practical applications across various fields. The main contributions of this review include: (1) presenting a comprehensive summary framework that systematically categorizes and compares different technological approaches, including symbolic generation, audio generation, and hybrid models, helping readers better understand the full spectrum of technologies in the field; (2) offering an extensive survey of current literature, covering emerging topics such as multimodal datasets and emotion expression evaluation, providing a broad reference for related research; (3) conducting a detailed analysis of the practical impact of AI music generation in various application domains, particularly in real-time interaction and interdisciplinary applications, offering new perspectives and insights; (4) summarizing the existing challenges and limitations of music quality evaluation methods and proposing potential future research directions, aiming to promote the standardization and broader adoption of evaluation techniques. Through these innovative summaries and analyses, this paper serves as a comprehensive reference tool for researchers and practitioners in AI music generation, while also outlining future directions for the field.
Recent advancements in audio generation have been significantly propelled by the capabilities of Large Language Models (LLMs). The existing research on audio LLM has primarily focused on enhancing the architecture and scale of audio language models, as well as leveraging larger datasets, and generally, acoustic codecs, such as EnCodec, are used for audio tokenization. However, these codecs were originally designed for audio compression, which may lead to suboptimal performance in the context of audio LLM. Our research aims to address the shortcomings of current audio LLM codecs, particularly their challenges in maintaining semantic integrity in generated audio. For instance, existing methods like VALL-E, which condition acoustic token generation on text transcriptions, often suffer from content inaccuracies and elevated word error rates (WER) due to semantic misinterpretations of acoustic tokens, resulting in word skipping and errors. To overcome these issues, we propose a straightforward yet effective approach called X-Codec. X-Codec incorporates semantic features from a pre-trained semantic encoder before the Residual Vector Quantization (RVQ) stage and introduces a semantic reconstruction loss after RVQ. By enhancing the semantic ability of the codec, X-Codec significantly reduces WER in speech synthesis tasks and extends these benefits to non-speech applications, including music and sound generation. Our experiments in text-to-speech, music continuation, and text-to-sound tasks demonstrate that integrating semantic information substantially improves the overall performance of language models in audio generation. Our code and demo are available (Demo: https://x-codec-audio.github.io Code: https://github.com/zhenye234/xcodec)
Foley sound synthesis is crucial for multimedia production, enhancing user experience by synchronizing audio and video both temporally and semantically. Recent studies on automating this labor-intensive process through video-to-sound generation face significant challenges. Systems lacking explicit temporal features suffer from poor alignment and controllability, while timestamp-based models require costly and subjective human annotation. We propose Video-Foley, a video-to-sound system using Root Mean Square (RMS) as an intuitive condition with semantic timbre prompts (audio or text). RMS, a frame-level intensity envelope closely related to audio semantics, acts as a temporal event feature to guide audio generation from video. The annotation-free self-supervised learning framework consists of two stages, Video2RMS and RMS2Sound, incorporating novel ideas including RMS discretization and RMS-ControlNet with a pretrained text-to-audio model. Our extensive evaluation shows that Video-Foley achieves state-of-the-art performance in audio-visual alignment and controllability for sound timing, intensity, timbre, and nuance. Source code, model weights and demos are available on our companion website. (https://jnwnlee.github.io/video-foley-demo)
Audio generation has achieved remarkable progress with the advance of sophisticated generative models, such as diffusion models (DMs) and autoregressive (AR) models. However, due to the naturally significant sequence length of audio, the efficiency of audio generation remains an essential issue to be addressed, especially for AR models that are incorporated in large language models (LLMs). In this paper, we analyze the token length of audio tokenization and propose a novel \textbf{S}cale-level \textbf{A}udio \textbf{T}okenizer (SAT), with improved residual quantization. Based on SAT, a scale-level \textbf{A}coustic \textbf{A}uto\textbf{R}egressive (AAR) modeling framework is further proposed, which shifts the next-token AR prediction to next-scale AR prediction, significantly reducing the training cost and inference time. To validate the effectiveness of the proposed approach, we comprehensively analyze design choices and demonstrate the proposed AAR framework achieves a remarkable \textbf{35}$\times$ faster inference speed and +\textbf{1.33} Fr\'echet Audio Distance (FAD) against baselines on the AudioSet benchmark. Code: \url{https://github.com/qiuk2/AAR}.
Deep learning (DL) models based on the transformer architecture have revolutionized many DL applications such as large language models (LLMs), vision transformers, audio generation, and time series prediction. Much of this progress has been fueled by distributed training, yet distributed communication remains a substantial bottleneck to training progress. This paper examines the communication behavior of transformer models - that is, how different parallelism schemes used in multi-node/multi-GPU DL Training communicate data in the context of transformers. We use GPT-based language models as a case study of the transformer architecture due to their ubiquity. We validate the empirical results obtained from our communication logs using analytical models. At a high level, our analysis reveals a need to optimize small message point-to-point communication further, correlations between sequence length, per-GPU throughput, model size, and optimizations used, and where to potentially guide further optimizations in framework and HPC middleware design and optimization.
Diffusion models (DM) have become fundamental components of generative models, excelling across various domains such as image creation, audio generation, and complex data interpolation. Signal-to-Noise diffusion models constitute a diverse family covering most state-of-the-art diffusion models. While there have been several attempts to study Signal-to-Noise (S2N) diffusion models from various perspectives, there remains a need for a comprehensive study connecting different viewpoints and exploring new perspectives. In this study, we offer a comprehensive perspective on noise schedulers, examining their role through the lens of the signal-to-noise ratio (SNR) and its connections to information theory. Building upon this framework, we have developed a generalized backward equation to enhance the performance of the inference process.