We introduce Noise2Music, where a series of diffusion models is trained to generate high-quality 30-second music clips from text prompts. Two types of diffusion models, a generator model, which generates an intermediate representation conditioned on text, and a cascader model, which generates high-fidelity audio conditioned on the intermediate representation and possibly the text, are trained and utilized in succession to generate high-fidelity music. We explore two options for the intermediate representation, one using a spectrogram and the other using audio with lower fidelity. We find that the generated audio is not only able to faithfully reflect key elements of the text prompt such as genre, tempo, instruments, mood, and era, but goes beyond to ground fine-grained semantics of the prompt. Pretrained large language models play a key role in this story -- they are used to generate paired text for the audio of the training set and to extract embeddings of the text prompts ingested by the diffusion models. Generated examples: https://google-research.github.io/noise2music
Foundation models (FMs), that are trained on broad data at scale and are adaptable to a wide range of downstream tasks, have brought large interest in the research community. Benefiting from the diverse data sources such as different modalities, languages and application domains, foundation models have demonstrated strong generalization and knowledge transfer capabilities. In this paper, we present a pioneering study towards building an efficient solution for FM-based speech recognition systems. We adopt the recently developed self-supervised BEST-RQ for pretraining, and propose the joint finetuning with both source and unsupervised target domain data using JUST Hydra. The FM encoder adapter and decoder are then finetuned to the target domain with a small amount of supervised in-domain data. On a large-scale YouTube and Voice Search task, our method is shown to be both data and model parameter efficient. It achieves the same quality with only 21.6M supervised in-domain data and 130.8M finetuned parameters, compared to the 731.1M model trained from scratch on additional 300M supervised in-domain data.
Recently, the concept of holographic multiple-input multiple-output (MIMO) is emerging as one of the promising technologies beyond massive MIMO. Many challenges need to be addressed to bring this novel idea into practice, including electromagnetic (EM)-compliant channel modeling and accurate performance evaluation. In this paper, an EM-compliant channel model is proposed for the holographic MIMO systems, which is able to model both the characteristics of the propagation channel and the non-ideal factors caused by mutual coupling at the transceivers, including the antenna pattern distortion and the decrease of antenna efficiency. Based on the proposed channel model, a more realistic performance evaluation is conducted to show the performance of the holographic MIMO system in both the single-user and the multi-user scenarios. Key challenges and future research directions are further provided based on the theoretical analyses and numerical results.
Most research on task oriented dialog modeling is based on written text input. However, users interact with practical dialog systems often using speech as input. Typically, systems convert speech into text using an Automatic Speech Recognition (ASR) system, introducing errors. Furthermore, these systems do not address the differences in written and spoken language. The research on this topic is stymied by the lack of a public corpus. Motivated by these considerations, our goal in hosting the speech-aware dialog state tracking challenge was to create a public corpus or task which can be used to investigate the performance gap between the written and spoken forms of input, develop models that could alleviate this gap, and establish whether Text-to-Speech-based (TTS) systems is a reasonable surrogate to the more-labor intensive human data collection. We created three spoken versions of the popular written-domain MultiWoz task -- (a) TTS-Verbatim: written user inputs were converted into speech waveforms using a TTS system, (b) Human-Verbatim: humans spoke the user inputs verbatim, and (c) Human-paraphrased: humans paraphrased the user inputs. Additionally, we provided different forms of ASR output to encourage wider participation from teams that may not have access to state-of-the-art ASR systems. These included ASR transcripts, word time stamps, and latent representations of the audio (audio encoder outputs). In this paper, we describe the corpus, report results from participating teams, provide preliminary analyses of their results, and summarize the current state-of-the-art in this domain.
Although deep neural networks have shown well-performance in various tasks, the poor interpretability of the models is always criticized. In the paper, we propose a new interpretable neural network method, by embedding neurons into the semantic space to extract their intrinsic global semantics. In contrast to previous methods that probe latent knowledge inside the model, the proposed semantic vector externalizes the latent knowledge to static knowledge, which is easy to exploit. Specifically, we assume that neurons with similar activation are of similar semantic information. Afterwards, semantic vectors are optimized by continuously aligning activation similarity and semantic vector similarity during the training of the neural network. The visualization of semantic vectors allows for a qualitative explanation of the neural network. Moreover, we assess the static knowledge quantitatively by knowledge distillation tasks. Empirical experiments of visualization show that semantic vectors describe neuron activation semantics well. Without the sample-by-sample guidance from the teacher model, static knowledge distillation exhibit comparable or even superior performance with existing relation-based knowledge distillation methods.
We propose a novel method to accelerate training and inference process of recurrent neural network transducer (RNN-T) based on the guidance from a co-trained connectionist temporal classification (CTC) model. We made a key assumption that if an encoder embedding frame is classified as a blank frame by the CTC model, it is likely that this frame will be aligned to blank for all the partial alignments or hypotheses in RNN-T and it can be discarded from the decoder input. We also show that this frame reduction operation can be applied in the middle of the encoder, which result in significant speed up for the training and inference in RNN-T. We further show that the CTC alignment, a by-product of the CTC decoder, can also be used to perform lattice reduction for RNN-T during training. Our method is evaluated on the Librispeech and SpeechStew tasks. We demonstrate that the proposed method is able to accelerate the RNN-T inference by 2.2 times with similar or slightly better word error rates (WER).
Existing multimodal tasks mostly target at the complete input modality setting, i.e., each modality is either complete or completely missing in both training and test sets. However, the randomly missing situations have still been underexplored. In this paper, we present a novel approach named MM-Align to address the missing-modality inference problem. Concretely, we propose 1) an alignment dynamics learning module based on the theory of optimal transport (OT) for indirect missing data imputation; 2) a denoising training algorithm to simultaneously enhance the imputation results and backbone network performance. Compared with previous methods which devote to reconstructing the missing inputs, MM-Align learns to capture and imitate the alignment dynamics between modality sequences. Results of comprehensive experiments on three datasets covering two multimodal tasks empirically demonstrate that our method can perform more accurate and faster inference and relieve overfitting under various missing conditions.
Self-training methods have been explored in recent years and have exhibited great performance in improving semi-supervised learning. This work presents a Simple instance-Adaptive self-Training method (SAT) for semi-supervised text classification. SAT first generates two augmented views for each unlabeled data and then trains a meta-learner to automatically identify the relative strength of augmentations based on the similarity between the original view and the augmented views. The weakly-augmented view is fed to the model to produce a pseudo-label and the strongly-augmented view is used to train the model to predict the same pseudo-label. We conducted extensive experiments and analyses on three text classification datasets and found that with varying sizes of labeled training data, SAT consistently shows competitive performance compared to existing semi-supervised learning methods. Our code can be found at \url{https://github.com/declare-lab/SAT.git}.
With the boom of e-commerce, Multimodal Review Helpfulness Prediction (MRHP), which aims to sort product reviews according to the predicted helpfulness scores has become a research hotspot. Previous work on this task focuses on attention-based modality fusion, information integration, and relation modeling, which primarily exposes the following drawbacks: 1) the model may fail to capture the really essential information due to its indiscriminate attention formulation; 2) lack appropriate modeling methods that take full advantage of correlation among provided data. In this paper, we propose SANCL: Selective Attention and Natural Contrastive Learning for MRHP. SANCL adopts a probe-based strategy to enforce high attention weights on the regions of greater significance. It also constructs a contrastive learning framework based on natural matching properties in the dataset. Experimental results on two benchmark datasets with three categories show that SANCL achieves state-of-the-art baseline performance with lower memory consumption.
This paper proposes a simple yet effective interpolation-based data augmentation approach termed DoubleMix, to improve the robustness of models in text classification. DoubleMix first leverages a couple of simple augmentation operations to generate several perturbed samples for each training data, and then uses the perturbed data and original data to carry out a two-step interpolation in the hidden space of neural models. Concretely, it first mixes up the perturbed data to a synthetic sample and then mixes up the original data and the synthetic perturbed data. DoubleMix enhances models' robustness by learning the "shifted" features in hidden space. On six text classification benchmark datasets, our approach outperforms several popular text augmentation methods including token-level, sentence-level, and hidden-level data augmentation techniques. Also, experiments in low-resource settings show our approach consistently improves models' performance when the training data is scarce. Extensive ablation studies and case studies confirm that each component of our approach contributes to the final performance and show that our approach exhibits superior performance on challenging counterexamples. Additionally, visual analysis shows that text features generated by our approach are highly interpretable. Our code for this paper can be found at https://github.com/declare-lab/DoubleMix.git.