Jump to Content
Rif A. Saurous

Rif A. Saurous

Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    Sequential Monte Carlo Learning for Time Series Structure Discovery
    Feras Saad
    Vikash Mansinghka
    Proceedings of the 40th International Conference on Machine Learning (2023), pp. 29473-29489
    Preview abstract This paper presents a new approach to automatically discovering accurate models of complex time series data. Working within a Bayesian nonparametric prior over a symbolic space of Gaussian process time series models, we present a novel structure learning algorithm that integrates sequential Monte Carlo (SMC) and involutive MCMC for highly effective posterior inference. Our method can be used both in "online'' settings, where new data is incorporated sequentially in time, and in "offline'' settings, by using nested subsets of historical data to anneal the posterior. Empirical measurements on a variety of real-world time series show that our method can deliver 10x--100x runtime speedups over previous MCMC and greedy-search structure learning algorithms for the same model family. We use our method to perform the first large-scale evaluation of Gaussian process time series structure learning on a widely used benchmark of 1,428 monthly econometric datasets, showing that our method discovers sensible models that deliver more accurate point forecasts and interval forecasts over multiple horizons as compared to prominent statistical and neural baselines that struggle on this challenging data. View details
    Preview abstract Humans do not acquire perceptual abilities like we train machines. While machine learning algorithms typically operate on large collections of randomly-chosen, explicitly-labeled examples, human acquisition relies far greater on multimodal unsupervised learning (as infants) and active learning (as children). With this motivation, we present a learning framework for sound representation and recognition that combines (i) a self-supervised objective based on a general notion of unimodal and cross-modal coincidence, (ii) a novel clustering objective that reflects our need to impose categorical structure on our experiences, and (iii) a cluster-based active learning procedure that solicits targeted weak supervision to consolidate hypothesized categories into relevant semantic classes. By jointly training a single sound embedding/clustering/classification network according to these criteria, we achieve a new state-of-the-art unsupervised audio representation and demonstrate up to 20-fold reduction in labels required to reach a desired classification performance. View details
    Preview abstract We explore content-based representation learning strategies tailored for large-scale, uncurated music collections that afford only weak supervision through unstructured natural language metadata and co-listen statistics. At the core is a hybrid training scheme that uses classification and metric learning losses to incorporate both metadata-derived text labels and aggregate co-listen supervisory signals into a single convolutional model. The resulting joint text and audio content embedding defines a similarity metric and supports prediction of semantic text labels using a vocabulary of unprecedented granularity, which we refine using a novel word-sense disambiguation procedure. As input to simple classifier architectures, our representation achieves state-of-the-art performance on two music tagging benchmarks. View details
    Automatically batching control-intensive programs for modern accelerators
    Alexey Radul
    Dougal Maclaurin
    Third Conference on Systems and Machine Learning, Austin, TX (2020)
    Preview abstract We present a general approach to batching arbitrary computations for GPU and TPU accelerators. We demonstrate the effectiveness of our method with orders-of-magnitude speedups on the No U-Turn Sampler (NUTS), a workhorse algorithm in Bayesian statistics. The central challenge of batching NUTS and other Markov chain Monte Carlo algorithms is data-dependent control flow and recursion. We overcome this by mechanically transforming a single-example implementation into a form that explicitly tracks the current program point for each batch member, and only steps forward those in the same place. We present two different batching algorithms: a simpler, previously published one that inherits recursion from the host Python, and a more complex, novel one that implmenents recursion directly and can batch across it. We implement these batching methods as a general program transformation on Python source. Both the batching system and the NUTS implementation presented here are available as part of the popular TensorFlow Probability software package. View details
    Estimating the Changing Infection Rate of COVID-19 Using Bayesian Models of Mobility
    Xue Ben
    Alexander Nicholas D'Amour
    Shawn O'Banion
    medRxiv, vol. https://www.medrxiv.org/content/10.1101/2020.08.06.20169664v1.full (2020)
    Preview abstract In order to prepare for and control the continued spread of the COVID-19 pandemic while minimizing its economic impact, the world needs to be able to estimate and predict COVID-19’s spread. Unfortunately, we cannot directly observe the prevalence or growth rate of COVID-19; these must be inferred using some kind of model. We propose a hierarchical Bayesian extension to the classic susceptible-exposed-infected-removed (SEIR) compartmental model that adds compartments to account for isolation and death and allows the infection rate to vary as a function of both mobility data collected from mobile phones and a latent time-varying factor that accounts for changes in behavior not captured by mobility data. Since confirmed-case data is unreliable, we infer the model’s parameters conditioned on deaths data. We replace the exponential-waiting-time assumption of classic compartmental models with Erlang distributions, which allows for a more realistic model of the long lag between exposure and death. The mobility data gives us a leading indicator that can quickly detect changes in the pandemic’s local growth rate and forecast changes in death rates weeks ahead of time. This is an analysis of observational data, so any causal interpretations of the model's inferences should be treated as suggestive at best; nonetheless, the model’s inferred relationship between different kinds of trips and the infection rate do suggest some possible hypotheses about what kinds of activities might contribute most to COVID-19’s spread. View details
    Differentiable Consistency Constraints for Improved Deep Speech Enhancement
    Jeremy Thorpe
    Michael Chinen
    IEEE International Conference on Acoustics, Speech, and Signal Processing (2019)
    Preview abstract In recent years, deep networks have led to dramatic improvements in speech enhancement by framing it as a data-driven pattern recognition problem. In many modern enhancement systems, large amounts of data are used to train a deep network to estimate masks for complex-valued short-time Fourier transforms (STFTs) to suppress noise and preserve speech. However, current masking approaches often neglect two important constraints: STFT consistency and mixture consistency. Without STFT consistency, the system’s output is not necessarily the STFT of a time-domain signal, and without mixture consistency, the sum of the estimated sources does not necessarily equal the input mixture. Furthermore, the only previous approaches that apply mixture consistency use real-valued masks; mixture consistency has been ignored for complex-valued masks. In this paper, we show that STFT consistency and mixture consistency can be jointly imposed by adding simple differentiable projection layers to the enhancement network. These layers are compatible with real or complex-valued masks. Using both of these constraints with complex-valued masks provides a 0.7 dB increase in scale-invariant signal-to-distortion ratio (SI-SDR) on a large dataset of speech corrupted by a wide variety of nonstationary noise across a range of input SNRs. View details
    Neumann Optimizer: A Practical Optimizer for Deep Neural Networks
    Shankar Krishnan
    Ying Xiao
    International Conference on Learning Representations (ICLR) (2018)
    Preview abstract Progress in deep learning is slowed by the days or weeks it takes to train large models. The natural solution of using more hardware is limited by diminishing returns, and leads to inefficient use of additional resources. In this paper, we present a large batch, stochastic optimization algorithm that is both faster than widely used algorithms for fixed amounts of computation, and is also able to scale up substantially better as more computational resources become available. Our algorithm implicitly computes the inverse hessian of each mini-batch to produce descent directions. We demonstrate the effectiveness of our algorithm by successfully training large ImageNet models (Inception V3, Resnet-50, Resnet-101 and Inception-Resnet) with mini-batch sizes of up to 32000 with no loss in validation error relative to current baselines, and no increase in the total number of steps. At smaller mini-batch sizes, our optimizer improves the validation error in these models by 0.8-0.9%. Alternatively, we can trade off this accuracy to reduce the number of training steps needed by roughly 10-30%. Our work is practical and easily usable by others -- only one hyperparameter (learning rate) needs tuning, and furthermore, the algorithm is as computationally cheap as the commonly used adam optimizer. View details
    Preview abstract Even in the absence of any explicit semantic annotation, vast collections of audio recordings provide valuable information for learning the categorical structure of sounds. We consider several class-agnostic semantic constraints that apply to unlabeled nonspeech audio: (i) noise and translations in time do not change the underlying sound category, (ii) a mixture of two sound events inherits the categories of the constituents, and (iii) the categories of events in close temporal proximity are likely to be the same or related. Without labels to ground them, these constraints are incompatible with classification loss functions. However, they may still be leveraged to identify geometric inequalities needed for triplet loss-based training of convolutional neural networks. The result is low-dimensional embeddings of the input spectrograms that recover 41% and 84% of the performance of their fully-supervised counterparts when applied to downstream query-by-example sound retrieval and sound event classification tasks, respectively. Moreover, in limited-supervision settings, our unsupervised embeddings double the state-of-the-art classification performance. View details
    Preview abstract Inspired by recent work on neural network image generation which rely on backpropagation towards the network inputs, we present a proof-of-concept system for speech texture synthesis and voice conversion based on two mechanisms: approximate inversion of the representation learned by a speech recognition neural network, and on matching statistics of neuron activations between different source and target utterances. Similar to image texture synthesis and neural style transfer, the system works by optimizing a cost function with respect to the input waveform samples. To this end we use a differentiable mel-filterbank feature extraction pipeline and train a convolutional CTC speech recognition network. Our system is able to extract speaker characteristics from very limited amounts of target speaker data, as little as a few seconds, and can be used to generate realistic speech babble or reconstruct an utterance in a different voice. View details
    Preview abstract In this paper, we present a novel system that separates the voice of a target speaker from multi-speaker signals, by making use of a reference signal from the target speaker. We achieve this by training two separate neural networks: (1) A speaker recognition network that produces speaker-discriminative embeddings; (2) A spectrogram masking network that takes both noisy spectrogram and speaker embedding as input, and produces a mask. Our system significantly reduces the speech recognition WER on multi-speaker signals, with minimal WER degradation on single-speaker signals. View details
    Simple, Distributed, and Accelerated Probabilistic Programming
    Dustin Tran
    Dave Moore
    Christopher Gordon Suter
    Alexey Radul
    Matthew Johnson
    NeurIPS (2018)
    Preview abstract We describe Edward2, a low-level probabilistic programming language. Edward2 distills the core of probabilistic programming down to a single abstraction—the random variable. By blurring the line between model and computation, Edward2 enables numerous applications not shown before: a model-parallel variational auto-encoder (VAE) with tensor processing units (TPUs); a data-parallel autoregressive model (Image Transformer) with TPUs; and multi-GPU No-U-Turn Sampler (NUTS). Edward2 achieves an optimal linear speedup from 4 to 256 TPUs. With VAEs, Edward2 sees up to a 20x speedup on TPUs over Pyro and Edward on GPUs; with Bayesian neural networks, Edward2 sees up to a 51x speedup. With NUTS, Edward2 sees a 20x speedup on GPUs over Stan and 7x over PyMC3. View details
    Preview abstract We present an extension to the Tacotron speech synthesis architecture that learns a latent embedding space of prosody, derived from a reference acoustic representation containing the desired prosody. We show that conditioning Tacotron on this learned embedding space results in synthesized audio that matches the reference signal’s prosody with fine time detail. We define several quantitative and subjective metrics for evaluating prosody transfer, and report results and audio samples from a single-speaker and 44-speaker Tacotron model on a prosody transfer task. View details
    Natural TTS Synthesis By Conditioning WaveNet On Mel Spectrogram Predictions
    Jonathan Shen
    Ruoming Pang
    Mike Schuster
    Navdeep Jaitly
    Zongheng Yang
    Yu Zhang
    Yuxuan Wang
    Yannis Agiomyrgiannakis
    ICASSP (2018)
    Preview abstract This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinion score (MOS) of 4.53 comparable to a MOS of 4.58 for professionally recorded speech. To validate our design choices, we present ablation studies of key components of our system and evaluate the impact of using mel spectrograms as the input to WaveNet instead of linguistic, duration, and F0 features. We further demonstrate that using a compact acoustic intermediate representation enables significant simplification of the WaveNet architecture. View details
    Fixing a Broken ELBO
    Ben Poole
    Ian Fischer
    Josh Dillon
    Proceedings of the 35th International Conference on Machine Learning, PMLR, Stockholmsmässan, Stockholm Sweden (2018), pp. 159-168
    Preview abstract Recent work in unsupervised representation learning has focused on learning deep directed latent variable models. Fitting these models by maximizing the marginal likelihood or evidence is typically intractable, thus a common approximation is to maximize the evidence lower bound (ELBO) instead. However, maximum likelihood training (whether exact or approximate) does not necessarily result in a good latent representation, as we demonstrate both theoretically and empirically. In particular, we derive variational lower and upper bounds on the mutual information between the input and the latent variable, and use these bounds to derive a rate-distortion curve that characterizes the tradeoff between compression and reconstruction accuracy. Using this framework, we demonstrate that there is a family of models with identical ELBO, but different quantitative and qualitative characteristics. Our framework also suggests a simple new method to ensure that latent variable models with powerful stochastic decoders do not ignore their latent code. View details
    Jeremy Thorpe
    Michael Chinen
    Proceedings of the 16th International Workshop on Acoustic Signal Enhancement (2018)
    Preview abstract We explore a variety of configurations of neural networks for one- and two-channel spectrogram-mask-based speech enhancement. Our best model improves on state-of-the-art performance on the CHiME2 speech enhancement task. We examine trade-offs among non-causal lookahead, compute work, and parameter count versus enhancement performance and find that zero-lookahead models can achieve, on average, only 0.5 dB worse performance than our best bidirectional model. Further, we find that 200 milliseconds of lookahead is sufficient to achieve performance within about 0.2 dB from our best bidirectional model. View details
    Preview abstract In this work, we propose “global style tokens”(GSTs), a bank of embeddings that are jointly trained within Tacotron, a state-of-the-art end-to-end speech synthesis system. The embeddings are trained in a completely unsupervised manner, and yet learn to model a large range of acoustic expressiveness. GSTs lead to a rich set of surprising results. The soft interpretable “labels” they generate can be used to control synthesis in novel ways, such as varying speed and modifying speak-ing style – independently of the text content. The labels can also be used for style transfer, replicating the speaking style of one “seed” phrase across an entire long-form text corpus. Perhaps most surprisingly, when trained on noisy, unlabelled found data, GSTs learn to factorize noise and speaker identity, providing a path towards highly scaleable but robust speech synthesis. View details
    TensorFlow Distributions
    Josh Dillon
    Dustin Tran
    Eugene Brevdo
    Dave Moore
    Workshop on Probabilistic Programming Languages, Semantics, and Systems (PPS 2018) (2017)
    Preview abstract The TensorFlow Distributions library implements a vision of probability theory adapted to the modern deep-learning paradigm of end-to-end differentiable computation. Building on two basic abstractions, it offers flexible building blocks for probabilistic computation. Distributions provide fast, numerically stable methods for generating samples and computing statistics, e.g., log density. Bijectors provide composable volume-tracking transformations with automatic caching. Together these enable modular construction of high dimensional distributions and transformations not possible with previous libraries (e.g., pixelCNNs, autoregressive flows, and reversible residual networks). They are the workhorse behind deep probabilistic programming systems like Edward and empower fast black-box inference in probabilistic models built on deep-network components. TensorFlow Distributions has proven an important part of the TensorFlow toolkit within Google and in the broader deep learning community. View details
    Uncovering Latent Style Factors for Expressive Speech Synthesis
    Yuxuan Wang
    Ying Xiao
    Joel Shor
    NIPS Workshop on Machine Learning for Audio Signal Processing (ML4Audio) (2017) (to appear)
    Preview abstract Prosodic modeling is a core problem in speech synthesis. The key challenge is producing desirable prosody from textual input containing only phonetic information. In this preliminary study, we introduce the concept of "style tokens" in Tacotron, a recently proposed end-to-end neural speech synthesis model. Using style tokens, we aim to extract independent prosodic styles from training data. We show that without annotation data or an explicit supervision signal, our approach can automatically learn a variety of prosodic variations in a purely data-driven way. Importantly, each style token corresponds to a fixed style factor regardless of the given text sequence. As a result, we can control the prosodic style of synthetic speech in a somewhat predictable and globally consistent way. View details
    Towards Learning Semantic Audio Representations from Unlabeled Data
    Ratheet Pandya
    Dan Ellis
    Jiayang Liu
    NIPS Workshop on Machine Learning for Audio Signal Processing (ML4Audio) (2017) (to appear)
    Preview abstract Our goal is to learn semantically structured audio representations without relying on categorically labeled data. We consider several class-agnostic semantic constraints that are inherent to non-speech audio: (i) sound categories are invariant to additive noise and translations in time, (ii) mixtures of two sound events inherit the categories of the constituents, and (iii) the categories of events in close temporal proximity in a single recording are likely to be the same or related. We apply these invariants in the service of sampling training data for triplet-loss embedding models using a large unlabeled dataset of YouTube soundtracks. The resulting low-dimensional representations provide both greatly improved query-by-example retrieval performance and reduced labeled data and model complexity requirements for supervised sound classification. View details
    Preview abstract A text-to-speech synthesis system typically consists of multiple stages, such as a text analysis frontend, an acoustic model and an audio synthesis module. Building these components often requires extensive domain expertise and may contain brittle design choices. In this paper, we present Tacotron, an end-to-end generative text-to-speech model that synthesizes speech directly from characters. Given (text, audio) pairs, the model can be trained completely from scratch with random initialization. We present several key techniques to make the sequence-to-sequence framework perform well for this challenging task. Tacotron achieves a 3.82 subjective 5-scale mean opinion score on US English, outperforming a production parametric system in terms of naturalness. In addition, since Tacotron generates speech at the frame level, it's substantially faster than sample-level autoregressive methods. View details
    Preview abstract Robust and far-field speech recognition is critical to enable true hands-free communication. In far-field conditions, signals are attenuated due to distance. To improve robustness to loudness variation, we introduce a novel frontend called per-channel energy normalization (PCEN). The key ingredient of PCEN is the use of an automatic gain control based dynamic compression to replace the widely used static (such as log or root) compression. We evaluate PCEN on the keyword spotting task. On our large rerecorded noisy and far-field eval sets, we show that PCEN significantly improves recognition performance. Furthermore, we model PCEN as neural network layers and optimize high-dimensional PCEN parameters jointly with the keyword spotting acoustic model. The trained PCEN frontend demonstrates significant further improvements without increasing model complexity or inference-time cost. View details
    CNN Architectures for Large-Scale Audio Classification
    Daniel P. W. Ellis
    Jort F. Gemmeke
    Devin Platt
    Malcolm Slaney
    International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE (2017)
    Preview abstract Convolutional Neural Networks (CNNs) have proven very effective in image classification and have shown promise for audio classification. We apply various CNN architectures to audio and investigate their ability to classify videos with a very large scale data set of 70M training videos (5.24 million hours) with 30,871 labels. We examine fully connected Deep Neural Networks (DNNs), AlexNet [1], VGG [2], Inception [3], and ResNet [4]. We explore the effects of training with different sized subsets of the 70M training videos. Additionally we report the effect of training over different subsets of the 30,871 labels. While our dataset contains video-level labels, we are also interested in Acoustic Event Detection (AED) and train a classifier on embeddings learned from the video-level task on AudioSet [5]. We find that derivatives of image classification networks do well on our audio classification task, that increasing the number of labels we train on provides some improved performance over subsets of labels, that performance of models improves as we increase training set size, and that a model using embeddings learned from the video-level task do much better than a baseline on the AudioSet classification task. View details
    Preview abstract Modern retrieval systems are often driven by an underlying machine learning model. The goal of such systems is to identify and possibly rank the few most relevant items for a given query or context. Thus, such systems are typically evaluated using a ranking-based performance metric such as the area under the precision-recall curve, the Fβ score, precision at fixed recall, etc. Obviously, it is desirable to train such systems to optimize the metric of interest. In practice, due to the scalability limitations of existing approaches for optimizing such objectives, large-scale retrieval systems are instead trained to maximize classification accuracy, in the hope that performance as measured via the true objective will also be favorable. In this work we present a unified framework that, using straightforward building block bounds, allows for highly scalable optimization of a wide range of ranking-based objectives. We demonstrate the advantage of our approach on several real-life retrieval problems that are significantly larger than those considered in the literature, while achieving substantial improvement in performance over the accuracy-objective baseline. View details
    AutoMOS: Learning a non-intrusive assessor of naturalness-of-speech
    Yannis Agiomyrgiannakis
    NIPS 2016 End-to-end Learning for Speech and Audio Processing Workshop (to appear)
    Preview abstract Developers of text-to-speech synthesizers (TTS) often make use of human raters to assess the quality of synthesized speech. We demonstrate that we can model human raters' mean opinion scores (MOS) of synthesized speech using a deep recurrent neural network whose inputs consist solely of a raw waveform. Our best models provide utterance-level estimates of MOS only moderately inferior to sampled human ratings, as shown by Pearson and Spearman correlations. When multiple utterances are scored and averaged, a scenario common in synthesizer quality assessment, we achieve correlations comparable to those of human raters. This model has a number of applications, such as the ability to automatically explore the parameter space of a speech synthesizer without requiring a human-in-the-loop. We explore a method of probing what the models have learned. View details
    No Results Found