Mixture-of-experts (MoE) models have expanded from language modeling to automatic speech recognition (ASR). Traditional MoE methods, such as the Switch Transformer, route experts independently within each layer. Our analysis reveals that routers in most layers make expert choices that are not strongly correlated with the choices of the routers in other layers. To increase the cooperation between experts in different layers and encourage greater specialization, we use a shared router across different MoE layers. We call this model Omni-router Transformer. Extensive experiments on a large-scale pseudolabeled dataset and evaluations across 10 diverse, out-of-domain datasets demonstrate that the Omni-router Transformer is able to achieve lower training loss and consistently outperform dense and Switch Transformer models, reducing average word error rates by 11.2% and 8.2%, respectively, while promoting structured expert usage and improved robustness to diverse data.

The emergence of large language models (LLMs) has transformed spoken dialog systems, yet the optimal architecture for real-time on-device voice agents remains an open question. While end-to-end approaches promise theoretical advantages, cascaded systems (CSs) continue to outperform them in language understanding tasks, despite being constrained by sequential processing latency. In this work, we introduce ChipChat, a novel low-latency CS that overcomes traditional bottlenecks through architectural innovations and streaming optimizations. Our system integrates streaming (a) conversational speech recognition with mixture-of-experts, (b) state-action augmented LLM, (c) text-to-speech synthesis, (d) neural vocoder, and (e) speaker modeling. Implemented using MLX, ChipChat achieves subsecond response latency on a Mac Studio without dedicated GPUs, while preserving user privacy through complete on-device processing. Our work shows that strategically redesigned CSs can overcome their historical latency limitations, offering a promising path forward for practical voice-based AI agents.

Diffusion models have become the dominant approach for visual generation. They are trained by denoising a Markovian process which gradually adds noise to the input. We argue that the Markovian property limits the model’s ability to fully utilize the generation trajectory, leading to inefficiencies during training and inference. In this paper, we propose DART, a transformer-based model that unifies autoregressive (AR) and diffusion within a non-Markovian framework. DART iteratively denoises image patches spatially and spectrally using an AR model that has the same architecture as standard language models. DART does not rely on image quantization, which enables more effective image modeling while maintaining flexibility. Furthermore, DART seamlessly trains with both text and image data in a unified model. Our approach demonstrates competitive performance on class-conditioned and text-to-image generation tasks, offering a scalable, efficient alternative to traditional diffusion models. Through this unified framework, DART sets a new benchmark for scalable, high-quality image synthesis.

We present SWE-Gym, the first environment for training real-world software engineering (SWE) agents. SWE-Gym contains 2,438 real-world Python task instances, each comprising a codebase with an executable runtime environment, unit tests, and a task specified in natural language. We use SWE-Gym to train language model based SWE agents, achieving up to 19% absolute gains in resolve rate on the popular SWE-Bench Verified and Lite test sets. We also experiment with inference-time scaling through verifiers trained on agent trajectories sampled from SWE-Gym. When combined with our fine-tuned SWE agents, we achieve 32.0% and 26.0% on SWE-Bench Verified and Lite, respectively, reflecting a new state-of-the-art for open-weight SWE agents. To facilitate further research, we publicly release SWE-Gym, models, and agent trajectories.

Normalizing Flows (NFs) are likelihood-based models for continuous inputs. They have demonstrated promising results on both density estimation and generative modeling tasks, but have received relatively little attention in recent years. In this work, we demonstrate that NFs are more powerful than previously believed. We present TarFlow: a simple and scalable architecture that enables highly performant NF models. TarFlow can be thought of as a Transformer-based variant of Masked Autoregressive Flows (MAFs): it consists of a stack of autoregressive Transformer blocks on image patches, alternating the autoregression direction between layers. TarFlow is straightforward to train end-to-end, and capable of directly modeling and generating pixels. We also propose three key techniques to improve sample quality: Gaussian noise augmentation during training, a post training denoising procedure, and an effective guidance method for both class-conditional and unconditional settings. Putting these together, TarFlow sets new state-of-the-art results on likelihood estimation for images, beating the previous best methods by a large margin, and generates samples with quality and diversity comparable to diffusion models, for the first time with a stand-alone NF model. We make our code available at https://github.com/apple/ml-tarflow.

Discrete diffusion is a promising framework for modeling and generating discrete data. In this work, we present Target Concrete Score Matching (TCSM), a novel and versatile objective for training and fine-tuning discrete diffusion models. TCSM provides a general framework with broad applicability. It supports pre-training discrete diffusion models directly from data samples, and many existing discrete diffusion approaches naturally emerge as special cases of our more general TCSM framework. Furthermore, the same TCSM objective extends to post-training of discrete diffusion models, including fine-tuning using reward functions or preference data, and distillation of knowledge from pre-trained autoregressive models. These new capabilities stem from the core idea of TCSM, estimating the concrete score of the target distribution, which resides in the original (clean) data space. This allows seamless integration with reward functions and pre-trained models, which inherently only operate in the clean data space rather than the noisy intermediate spaces of diffusion processes. Our experiments on language modeling tasks demonstrate that TCSM matches or surpasses current methods. Additionally, TCSM is versatile, applicable to both pre-training and post-training scenarios, offering greater flexibility and sample efficiency.

Achieving human-level performance on real world tasks is a north star for the robotics community. We present the first learned robot agent that reaches amateur humanlevel performance in competitive table tennis. Table tennis is a physically demanding sport that takes humans years to master. We contribute (1) a hierarchical and modular policy architecture consisting of (i) low level controllers with their skill descriptors that model their capabilities and (ii) a high level controller that chooses the low level skills, (2) techniques for enabling zero-shot sim-to-real and curriculum building, including an iterative approach (train in sim, deploy in real), and (3) real time adaptation to unseen opponents. Policy performance was assessed through 29 robot vs. human matches of which the robot won 45 % (13/29). All humans were unseen players and their skill level varied from beginner to tournament level. Whilst the robot lost all matches vs. the most advanced players it won 100 % matches vs. beginners and 55 % matches vs. intermediate players, demonstrating solidly amateur humanlevel performance. Videos of the matches can be viewed here1.See sites https://google.com/view/competitive-robot-table-tennis.

Autoregressive models have driven remarkable progress in language modeling. Their foundational reliance on discrete tokens, unidirectional context, and single-pass decoding, while central to their success, also inspires the exploration of a design space that could offer new axes of modeling flexibility. In this work, we explore an alternative paradigm, shifting language modeling from a discrete token space to a continuous latent space. We propose a novel framework that employs transformer-based autoregressive normalizing flows to model these continuous representations. This approach unlocks substantial flexibility, enabling the construction of models that can capture global bi-directional context through stacked, alternating-direction autoregressive transformations, support block-wise generation with flexible token patch sizes, and facilitate a hierarchical multi-pass generation process. We further propose new mixture-based coupling transformations designed to capture complex dependencies within the latent space shaped by discrete data, and demonstrate theoretical connections to conventional discrete autoregressive models. Extensive experiments on language modeling benchmarks demonstrate strong likelihood performance and highlight the flexible modeling capabilities inherent in our framework.

Large language models (LLMs) are effective at answering questions that are clearly asked.However, when faced with ambiguous queries they can act unpredictably and produce incorrect outputs.This underscores the need for the development of intelligent agents capable of asking clarification questions to resolve ambiguities effectively.This capability requires complex understanding, state tracking, reasoning and planning over multiple conversational turns.However, directly measuring this can be challenging.In this paper, we offer a surrogate problem which assesses an LLMs's capability to deduce an entity unknown to itself, but revealed to a judge, by asking the judge a series of queries.This entity-deducing game can serve as an evaluation framework to probe the conversational reasoning and planning capabilities of language models.We systematically evaluate various LLMs and discover significant differences in their performance on this task.We find that strong LLMs like GPT-4 outperform human players by a large margin.We further employ Behavior Cloning (BC) to examine whether a weaker model is capable of imitating a stronger model and generalizing to data or domains, using only the demonstrations from a stronger model.We finally propose to use Reinforcement Learning to enhance reasoning and planning capacity of Vicuna models through episodes of game playing, which lead to significant performance improvement.We hope that this problem offers insights into how autonomous agents could be trained to behave more intelligently in ambiguous circumstances. Get MeshuggaThe band Play their song Do you mean Meshugana, the story set, or Meshuggah, the band?Do you want to know their information, or to play their album?Do you want something popular or unique from them?I am thinking of a movie that I cannot remember the name.Can you help me?Nope, it's like 10 years ago.I think so!Is this movie a drama movie?

Pratyush Maini, Skyler Seto, Richard Bai, David Grangier, Yizhe Zhang, Navdeep Jaitly. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Datasets that pair Knowledge Graphs (KG) and text together (KG-T) can be used to train forward and reverse neural models that generate text from KG and vice versa. However models trained on datasets where KG and text pairs are not equivalent can suffer from more hallucination and poorer recall. In this paper, we verify this empirically by generating datasets with different levels of noise and find that noisier datasets do indeed lead to more hallucination. We argue that the ability of forward and reverse models trained on a dataset to cyclically regenerate source KG or text is a proxy for the equivalence between the KG and the text in the dataset. Using cyclic evaluation we find that manually created WebNLG is much better than automatically created TeKGen and T-REx. Informed by these observations, we construct a new, improved dataset called LAGRANGE using heuristics meant to improve equivalence between KG and text and show the impact of each of the heuristics on cyclic evaluation. We also construct two synthetic datasets using large language models (LLMs), and observe that these are conducive to models that perform significantly well on cyclic generation of text, but less so on cyclic generation of KGs, probably because of a lack of a consistent underlying ontology.

Diffusion models are the de-facto approach for generating high-quality images and videos, but learning high-dimensional models remains a formidable task due to computational and optimization challenges. Existing methods often resort to training cascaded models in pixel space, or using a downsampled latent space of a separately trained auto-encoder. In this paper, we introduce Matryoshka Diffusion (MDM), an end-to-end framework for high-resolution image and video synthesis. We propose a diffusion process that denoises inputs at multiple resolutions jointly and uses a NestedUNet architecture where features and parameters for small-scale inputs are nested within those of large scales. In addition, MDM enables a progressive training schedule from lower to higher resolutions, which leads to significant improvements in optimization for high-resolution generation. We demonstrate the effectiveness of our approach on various benchmarks, including class-conditioned image generation, high-resolution text-to-image, and text-to-video applications. Remarkably, we can train a single pixel-space model at resolutions of up to 1024x1024 pixels, demonstrating strong zero-shot generalization using the CC12M dataset, which contains only 12 million images. Code and pre-trained checkpoints are released at https://github.com/apple/ml-mdm.

We present a novel way to predict molecular conformers through a simple formulation that sidesteps many of the heuristics of prior works and achieves state of the art results by using the advantages of scale. By training a diffusion generative model directly on 3D atomic positions without making assumptions about the explicit structure of molecules (e.g. modeling torsional angles) we are able to radically simplify structure learning, and make it trivial to scale up the model sizes. This model, called Molecular Conformer Fields (MCF), works by parameterizing conformer structures as functions that map elements from a molecular graph directly to their 3D location in space. This formulation allows us to boil down the essence of structure prediction to learning a distribution over functions. Experimental results show that scaling up the model capacity leads to large gains in generalization performance without enforcing inductive biases like rotational equivariance. MCF represents an advance in extending diffusion models to handle complex scientific problems in a conceptually simple, scalable and effective manner.

Large pretrained vision-language models like CLIP have shown promising generalization capability, but may struggle in specialized domains (e.g., satellite imagery) or fine-grained classification (e.g., car models) where the visual concepts are unseen or under-represented during pretraining. Prompt learning offers a parameter-efficient finetuning framework that can adapt CLIP to downstream tasks even when limited annotation data are available. In this paper, we improve prompt learning by distilling the textual knowledge from natural language prompts (either human- or LLM-generated) to provide rich priors for those under-represented concepts. We first obtain a prompt ``summary'' aligned to each input image via a learned prompt aggregator. Then we jointly train a prompt generator, optimized to produce a prompt embedding that stays close to the aggregated summary while minimizing task loss at the same time. We dub such prompt embedding as Aggregate-and-Adapted Prompt Embedding (AAPE). AAPE is shown to be able to generalize to different downstream data distributions and tasks, including vision-language understanding tasks (e.g., few-shot classification, VQA) and generation tasks (image captioning) where AAPE achieves competitive performance. We also show AAPE is particularly helpful to handle non-canonical and OOD examples. Furthermore, AAPE learning eliminates LLM-based inference cost as required by baselines, and scales better with data and LLM model size.

Diffusion models have emerged as a powerful tool for generating high-quality images from textual descriptions. Despite their successes, these models often exhibit limited diversity in the sampled images, particularly when sampling with a high classifier-free guidance weight. To address this issue, we present Kaleido, a novel approach that enhances the diversity of samples by incorporating autoregressive latent priors. Kaleido integrates an autoregressive language model that encodes the original caption and generates latent variables, serving as abstract and intermediary representations for guiding and facilitating the image generation process. In this paper, we explore a variety of discrete latent representations, including textual descriptions, detection bounding boxes, object blobs, and visual tokens. These representations diversify and enrich the input conditions to the diffusion models, enabling more diverse outputs. Our experimental results demonstrate that Kaleido effectively broadens the diversity of the generated image samples from a given textual description while maintaining high image quality. Furthermore, we show that Kaleido adheres closely to the guidance provided by the generated latent variables, demonstrating its capability to effectively control and direct the image generation process.

Fully-test-time adaptation (F-TTA) can mitigate performance loss due to distribution shifts between train and test data (1) without access to the training data, and (2) without knowledge of the model training procedure. In online F-TTA, a pre-trained model is adapted using a stream of test samples by minimizing a self-supervised objective, such as entropy minimization. However, models adapted with online using entropy minimization, are unstable especially in single sample settings, leading to degenerate solutions, and limiting the adoption of TTA inference strategies. Prior works identify noisy, or unreliable, samples as a cause of failure in online F-TTA. One solution is to ignore these samples, which can lead to bias in the update procedure, slow adaptation, and poor generalization. In this work, we present a general framework for improving robustness of F-TTA to these noisy samples, inspired by self-paced learning and robust loss functions. Our proposed approach, Robust Entropy Adaptive Loss Minimization (REALM), achieves better adaptation accuracy than previous approaches throughout the adaptation process on corruptions of CIFAR-10 and ImageNet-1K, demonstrating its effectiveness.

Self-training (ST) and self-supervised learning (SSL) methods have demonstrated strong improvements in automatic speech recognition (ASR). In spite of these advances, to the best of our knowledge, there is no analysis of how the composition of the labelled and unlabelled datasets used in these methods affects the results. In this work we aim to analyse the effect of number of speakers in the training data on a recent SSL algorithm (wav2vec 2.0), and a recent ST algorithm (slimIPL). We perform a systematic analysis on both labeled and unlabeled data by varying the number of speakers while keeping the number of hours fixed and vice versa. Our findings suggest that SSL requires a large amount of unlabeled data to produce high accuracy results, while ST requires a sufficient number of speakers in the labelled data, especially in the low-regime setting. In this manner these two approaches improve supervised learning in different regimes of data composition.

Autoregressive models for text sometimes generate repetitive and low-quality output because errors accumulate during the steps of generation. This issue is often attributed to exposure bias -- the difference between how a model is trained, and how it is used during inference. Denoising diffusion models provide an alternative approach in which a model can revisit and revise its output. However, they can be computationally expensive and prior efforts on text have led to models that produce less fluent output compared to autoregressive models, especially for longer text and paragraphs. In this paper, we propose PLANNER, a model that combines latent semantic diffusion with autoregressive generation, to generate fluent text while exercising global control over paragraphs. The model achieves this by combining an autoregressive "decoding" module with a "planning" module that uses latent diffusion to generate semantic paragraph embeddings in a coarse-to-fine manner. The proposed method is evaluated on various conditional generation tasks, and results on semantic generation, text completion and summarization show its effectiveness in generating high-quality long-form text in an efficient manner.

Self-attention mechanisms model long-range context by using pairwise attention between all input tokens. In doing so, they assume a fixed attention granularity defined by the individual tokens (e.g., text characters or image pixels), which may not be optimal for modeling complex dependencies at higher levels. In this paper, we propose ContextPool to address this problem by adapting the attention granularity for each token. Inspired by the success of ConvNets that are combined with pooling to capture long-range dependencies, we learn to pool neighboring features for each token before computing attention in a given attention layer. The pooling weights and support size are adaptively determined, allowing the pooled features to encode meaningful context with varying scale. We show that ContextPool makes attention models more expressive, achieving strong performance often with fewer layers and thus significantly reduced cost. Experiments validate that our ContextPool module, when plugged into transformer models, matches or surpasses state-of-the-art performance using less compute on several language and image benchmarks, outperforms recent works with learned context sizes or sparse attention patterns, and is also applicable to ConvNets for efficient feature learning.

Transformers \cite{transformer} have gained increasing popularity in a wide range of applications, including Natural Language Processing (NLP), Computer Vision and Speech Recognition, because of their powerful representational capacity. However, harnessing this representational capacity effectively requires a large amount of data, strong regularization, or both, to mitigate overfitting. Recently, the power of the Transformer has been unlocked by self-supervised pretraining strategies based on masked autoencoders which rely on reconstructing masked inputs, directly, or contrastively from unmasked content. This pretraining strategy which has been used in BERT models in NLP \cite{bert}, Wav2Vec models in Speech \cite{wv2v2} and, recently, in MAE models in Vision \cite{beit, mae}, forces the model to learn about relationships between the content in different parts of the input using autoencoding related objectives. In this paper, we propose a novel, but surprisingly simple alternative to content reconstruction – that of predicting locations from content, without providing positional information for it. Doing so requires the Transformer to understand the positional relationships between different parts of the input, from their content alone. This amounts to an efficient implementation where the pretext task is a classification problem among all possible positions for each input token. We experiment on both Vision and Speech benchmarks, where our approach brings improvements over strong supervised training baselines and is comparable to modern unsupervised/self-supervised pretraining methods. Our method also enables Transformers trained without position embeddings to outperform ones trained with full position information.

In recent years, all-neural end-to-end approaches have obtained state-of-the-art results on several challenging automatic speech recognition (ASR) tasks. However, most existing works focus on building ASR models where train and test data are drawn from the same domain. This results in poor generalization characteristics on mismatched-domains: e.g., end-to-end models trained on short segments perform poorly when evaluated on longer utterances. In this work, we analyze the generalization properties of streaming and non-streaming recurrent neural network transducer (RNN-T) based end-to-end models in order to identify model components that negatively affect generalization performance. We propose two solutions: combining multiple regularization techniques during training, and using dynamic overlapping inference. On a long-form YouTube test set, when the non-streaming RNN-T model is trained with shorter segments of data, the proposed combination improves word error rate (WER) from 22.3% to 14.8%; when the streaming RNN-T model trained on short Search queries, the proposed techniques improve WER on the YouTube set from 67.0% to 25.3%. Finally, when trained on Librispeech, we find that dynamic overlapping inference improves WER on YouTube from 99.8% to 33.0%.

This paper presents the Imputer, a neural sequence model that generates output sequences iteratively via imputations. The Imputer is an iterative generation model, requiring only a constant number of generation steps independent of the number of input or output tokens. The Imputer can be trained to approximately marginalize over all possible alignments between the input and output sequences, and all possible generation orders. We present a tractable dynamic programming training algorithm, which yields a lower bound on the log marginal likelihood. When applied to end-to-end speech recognition, the Imputer outperforms prior non-autoregressive models and achieves competitive results to autoregressive models. On LibriSpeech test-other, the Imputer achieves 11.1 WER, outperforming CTC at 13.0 WER and seq2seq at 12.5 WER.

We propose a model-free algorithm for learning efficient policies capable of returning table tennis balls by controlling robot joints at a rate of 100Hz. We demonstrate that evolutionary search (ES) methods acting on CNN-based policy architectures for non-visual inputs and convolving across time learn compact controllers leading to smooth motions. Furthermore, we show that with appropriately tuned curriculum learning on the task and rewards, policies are capable of developing multi-modal styles, specifically forehand and backhand stroke, whilst achieving 80% return rate on a wide range of ball throws. We observe that multi-modality does not require any architectural priors, such as multi-head architectures or hierarchical policies.

Attention-based encoder-decoder architectures such as Listen, Attend, and Spell (LAS), subsume the acoustic, pronunciation and language model components of a traditional automatic speech recognition (ASR) system into a single neural network. In previous work, we have shown that such architectures are comparable to state-of-the-art ASR systems on dictation tasks, but it was not clear if such architectures would be practical for more challenging tasks such as voice search. In this work, we explore a variety of structural and optimization improvements to our LAS model which significantly improve performance. On the structural side, we show that word piece models can be used instead of graphemes. We also introduce a multi-head attention architecture, which offers improvements over the commonly-used single-head attention. On the optimization side, we explore synchronous training, scheduled sampling, label smoothing, and minimum word error rate optimization, which are all shown to improve accuracy. We present results with a unidirectional LSTM encoder for streaming recognition. On a 12, 500 hour voice search task, we find that the proposed changes improve the WER from 9.2% to 5.6%, while the best conventional system achieves 6.7%; on a dictation task our model achieves a WER of 4.1% compared to 5% for the conventional system.

There has recently been significant interest in hard attention models for tasks such as object recognition, visual captioning and speech recognition. Hard attention offers benefits over soft attention such as decreased computational cost, but training hard attention models can be difficult because of the discrete latent variables they introduce. Previous work used REINFORCE to approach these issues, however, it suffers from high-variance gradient estimates, resulting in slow convergence. In this paper, we tackle the problem of learning hard attention for a sequential task using variational inference methods, specifically the recently introduced Variational Inference for Monte Carlo Objectives (VIMCO) and Neural Variational Inference (NVIL). Furthermore, we propose a novel baseline that adapts VIMCO to this setting. We demonstrate our method on a phoneme recognition task in clean and noisy environments and show that our method outperforms REINFORCE, with the difference being greater for a more complicated task.

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 time-domain 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 conditioning input to WaveNet instead of linguistic, duration, and F0 features. We further show that using this compact acoustic intermediate representation allows for a significant reduction in the size of the WaveNet architecture.

In this paper we document our experiences with developing speech recognition for medical transcription -a system that automatically transcribes doctor-patient conversations.Towards this goal, we built a system along two different methodological lines -a Connectionist Temporal Classification (CTC) phoneme based model and a Listen Attend and Spell (LAS) grapheme based model.To train these models we used a corpus of anonymized conversations representing approximately 14,000 hours of speech.Because of noisy transcripts and alignments in the corpus, a significant amount of effort was invested in data cleaning issues.We describe a two-stage strategy we followed for segmenting the data.The data cleanup and development of a matched language model was essential to the success of the CTC based models.The LAS based models, however were found to be resilient to alignment and transcript noise and did not require the use of language models.CTC models were able to achieve a word error rate of 20.1%, and the LAS models were able to achieve 18.3%.Our analysis shows that both models perform well on important medical utterances and therefore can be practical for transcribing medical conversations.

Sequence-to-sequence models with soft attention had significant success in machine translation, speech recognition, and question answering. Though capable and easy to use, they require that the entirety of the input sequence is available at the beginning of inference, an assumption that is not valid for instantaneous translation and speech recognition. To address this problem, we present a new method for solving sequence-to-sequence problems using hard online alignments instead of soft offline alignments. The online alignments model is able to start producing outputs without the need to first process the entire input sequence. A highly accurate online sequence-to-sequence model is useful because it can be used to build an accurate voice-based instantaneous translator. Our model uses hard binary stochastic decisions to select the timesteps at which outputs will be produced. The model is trained to produce these stochastic decisions using a standard policy gradient method. In our experiments, we show that this model achieves encouraging performance on TIMIT and Wall Street Journal (WSJ) speech recognition datasets.

Sequence-to-sequence models have shown success in end-to-end speech recognition. However these models have only used shallow acoustic encoder networks. In our work, we successively train very deep convolutional networks to add more expressive power and better generalization for end-to-end ASR models. We apply network-in-network principles, batch normalization, residual connections and convolutional LSTMs to build very deep recurrent and convolutional structures. Our models exploit the spectral structure in the feature space and add computational depth without overfitting issues. We experiment with the WSJ ASR task and achieve 10.5% word error rate without any dictionary or language model using a 15 layer deep network.

The recently proposed Sequence-to-Sequence (seq2seq) framework advocates replacing complex data processing pipelines, such as an entire automatic speech recognition system, with a single neural network trained in an end-to-end fashion.In this contribution, we analyse an attention-based seq2seq speech recognition system that directly transcribes recordings into characters.We observe two shortcomings: overconfidence in its predictions and a tendency to produce incomplete transcriptions when language models are used.We propose practical solutions to both problems achieving competitive speaker independent word error rates on the Wall Street Journal dataset: without separate language models we reach 10.6% WER, while together with a trigram language model, we reach 6.7% WER.

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 pairs, the model can be trained completely from scratch with random initialization.We present several key techniques to make the sequence-tosequence 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.

We present a recurrent encoder-decoder deep neural network architecture that directly translates speech in one language into text in another.The model does not explicitly transcribe the speech into text in the source language, nor does it require supervision from the ground truth source language transcription during training.We apply a slightly modified sequence-to-sequence with attention architecture that has previously been used for speech recognition and show that it can be repurposed for this more complex task, illustrating the power of attention-based models.A single model trained end-to-end obtains state-of-the-art performance on the Fisher Callhome Spanish-English speech translation task, outperforming a cascade of independently trained sequence-to-sequence speech recognition and machine translation models by 1.8 BLEU points on the Fisher test set.In addition, we find that making use of the training data in both languages by multi-task training sequence-to-sequence speech translation and recognition models with a shared encoder network can improve performance by a further 1.4 BLEU points.

We present Listen, Attend and Spell (LAS), a neural speech recognizer that transcribes speech utterances directly to characters without pronunciation models, HMMs or other components of traditional speech recognizers. In LAS, the neural network architecture subsumes the acoustic, pronunciation and language models making it not only an end-to-end trained system but an end-to-end model. In contrast to DNN-HMM, CTC and most other models, LAS makes no independence assumptions about the probability distribution of the output character sequences given the acoustic sequence. Our system has two components: a listener and a speller. The listener is a pyramidal recurrent network encoder that accepts filter bank spectra as inputs. The speller is an attention-based recurrent network decoder that emits each character conditioned on all previous characters, and the entire acoustic sequence. On a Google voice search task, LAS achieves a WER of 14.1% without a dictionary or an external language model and 10.3% with language model rescoring over the top 32 beams. In comparison, the state-of-the-art CLDNN-HMM model achieves a WER of 8.0% on the same set.

Sequence-to-sequence models have achieved impressive results on various tasks. However, they are unsuitable for tasks that require incremental predictions to be made as more data arrives or tasks that have long input sequences and output sequences. This is because they generate an output sequence conditioned on an entire input sequence. In this paper, we present a Neural Transducer that can make incremental predictions as more input arrives, without redoing the entire computation. Unlike sequence-to-sequence models, the Neural Transducer computes the next-step distribution conditioned on the partially observed input sequence and the partially generated sequence. At each time step, the transducer can decide to emit zero to many output symbols. The data can be processed using an encoder and presented as input to the transducer. The discrete decision to emit a symbol at every time step makes it difficult to learn with conventional backpropagation. It is however possible to train the transducer by using a dynamic programming algorithm to generate target discrete decisions. Our experiments show that the Neural Transducer works well in settings where it is required to produce output predictions as data come in. We also find that the Neural Transducer performs well for long sequences even when attention mechanisms are not used.

A key problem in structured output prediction is enabling direct optimization of the task reward function that matters for test evaluation. This paper presents a simple and computationally efficient method that incorporates task reward into maximum likelihood training. We establish a connection between maximum likelihood and regularized expected reward, showing that they are approximately equivalent in the vicinity of the optimal solution. Then we show how maximum likelihood can be generalized by optimizing the conditional probability of auxiliary outputs that are sampled proportional to their exponentiated scaled rewards. We apply this framework to optimize edit distance in the output space, by sampling from edited targets. Experiments on speech recognition and machine translation for neural sequence to sequence models show notable improvements over maximum likelihood baseline by simply sampling from target output augmentations.

Recurrent Neural Networks can be trained to produce sequences of tokens given some input, as exemplified by recent results in machine translation and image captioning. The current approach to training them consists of maximizing the likelihood of each token in the sequence given the current (recurrent) state and the previous token. At inference, the unknown previous token is then replaced by a token generated by the model itself. This discrepancy between training and inference can yield errors that can accumulate quickly along the generated sequence. We propose a curriculum learning strategy to gently change the training process from a fully guided scheme using the true previous token, towards a less guided scheme which mostly uses the generated token instead. Experiments on several sequence prediction tasks show that this approach yields significant improvements. Moreover, it was used successfully in our winning bid to the MSCOCO image captioning challenge, 2015.

We introduce a new neural architecture to learn the conditional probability of an output sequence with elements that arediscrete tokens corresponding to positions in an input sequence.Such problems cannot be trivially addressed by existent approaches such as sequence-to-sequence and Neural Turing Machines,because the number of target classes in eachstep of the output depends on the length of the input, which is variable.Problems such as sorting variable sized sequences, and various combinatorialoptimization problems belong to this class. Our model solvesthe problem of variable size output dictionaries using a recently proposedmechanism of neural attention. It differs from the previous attentionattempts in that, instead of using attention to blend hidden units of anencoder to a context vector at each decoder step, it uses attention asa pointer to select a member of the input sequence as the output. We call this architecture a Pointer Net (Ptr-Net).We show Ptr-Nets can be used to learn approximate solutions to threechallenging geometric problems -- finding planar convex hulls, computingDelaunay triangulations, and the planar Travelling Salesman Problem-- using training examples alone. Ptr-Nets not only improve oversequence-to-sequence with input attention, butalso allow us to generalize to variable size output dictionaries.We show that the learnt models generalize beyond the maximum lengthsthey were trained on. We hope our results on these taskswill encourage a broader exploration of neural learning for discreteproblems.

This paper presents a speech recognition system that directly transcribes audio data with text, without requiring an intermediate phonetic representation. The system is based on a combination of the deep bidirectional LSTM recurrent neural network architecture and the Connectionist Temporal Classification objective function. A modification to the objective function is introduced that trains the network to minimise the expectation of an arbitrary transcription loss function. This allows a direct optimisation of the word error rate, even in the absence of a lexicon or language model. The system achieves a word error rate of 27.3% on the Wall Street Journal corpus with no prior linguistic information, 21.9% with only a lexicon of allowed words, and 8.2% with a trigram language model. Combining the network with a baseline system further reduces the error rate to 6.7%.

We describe a simple but effective way of using multi-frame targets to improve the accuracy of Artificial Neural NetworkHidden Markov Model (ANN-HMM) hybrid systems. In this approach a Deep Neural Network (DNN) is trained to predict the forced-alignment state of multiple frames using a separate softmax unit for each of the frames. This is in contrast to the usual method of training a DNN to predict only the state of the central frame. By itself this is not sufficient to improve accuracy of the system significantly. However, if we average the predictions for each frame from the different contexts it is associated with we achieve state of the art results on TIMIT using a fully connected Deep Neural Network without convolutional architectures or dropout training. On a 14 hour subset of Wall Street Journal (WSJ) using a context dependent DNN-HMM system it leads to a relative improvement of 6.4% on the dev set (testdev93) and 9.3% on test set (test-eval92).

Deep Bidirectional LSTM (DBLSTM) recurrent neural networks have recently been shown to give state-of-the-art performance on the TIMIT speech database. However, the results in that work relied on recurrent-neural-network-specific objective functions, which are difficult to integrate with existing large vocabulary speech recognition systems. This paper investigates the use of DBLSTM as an acoustic model in a standard neural network-HMM hybrid system. We find that a DBLSTM-HMM hybrid gives equally good results on TIMIT as the previous work. It also outperforms both GMM and deep network benchmarks on a subset of the Wall Street Journal corpus. However the improvement in word error rate over the deep network is modest, despite a great increase in framelevel accuracy. We conclude that the hybrid approach with DBLSTM appears to be well suited for tasks where acoustic modelling predominates. Further investigation needs to be conducted to understand how to better leverage the improvements in frame-level accuracy towards better word error rates.

In this paper we show how we can discover non-linear features of frames of spectrograms using a novel autoencoder. The autoencoder uses a neural network encoder that predicts how a set of prototypes called templates need to be transformed to reconstruct the data, and a decoder that is a function that performs this operation of transforming prototypes and reconstructing the input. We demonstrate this method on spectrograms from the TIMIT database. The features are used in a Deep Neural Network Hidden Markov Model (DNN-HMM) hybrid system for automatic speech recognition. On the TIMIT monophone recognition task we were able to achieve gains of 0.5% over Mel log spectra, by augmenting traditional the spectra with the predicted transformation parameters. Further, using the recently discovered ‘dropout’ training, we were able to achieve a phone error rate (PER) of 17.9% on the dev set and 19.5% on the test set, which, to our knowledge is the best reported number on this task using a hybrid system.

BACKGROUND: MultiAlign is a free software tool that aligns multiple liquid chromatography-mass spectrometry datasets to one another by clustering mass and chromatographic elution features across datasets. Applicable to both label-free proteomics and metabolomics comparative analyses, the software can be operated in several modes. For example, clustered features can be matched to a reference database to identify analytes, used to generate abundance profiles, linked to tandem mass spectra based on parent precursor masses, and culled for targeted liquid chromatography-tandem mass spectrometric analysis. MultiAlign is also capable of tandem mass spectral clustering to describe proteome structure and find similarity in subsequent sample runs. RESULTS: MultiAlign was applied to two large proteomics datasets obtained from liquid chromatography-mass spectrometry analyses of environmental samples. Peptides in the datasets for a microbial community that had a known metagenome were identified by matching mass and elution time features to those in an established reference peptide database. Results compared favorably with those obtained using existing tools such as VIPER, but with the added benefit of being able to trace clusters of peptides across conditions to existing tandem mass spectra. MultiAlign was further applied to detect clusters across experimental samples derived from a reactor biomass community for which no metagenome was available. Several clusters were culled for further analysis to explore changes in the community structure. Lastly, MultiAlign was applied to liquid chromatography-mass spectrometry-based datasets obtained from a previously published study of wild type and mitochondrial fatty acid oxidation enzyme knockdown mutants of human hepatocarcinoma to demonstrate its utility for analyzing metabolomics datasets. CONCLUSION: MultiAlign is an efficient software package for finding similar analytes across multiple liquid chromatography-mass spectrometry feature maps, as demonstrated here for both proteomics and metabolomics experiments. The software is particularly useful for proteomic studies where little or no genomic context is known, such as with environmental proteomics.

The use of Deep Belief Networks (DBN) to pretrain Neural Networks has recently led to a resurgence in the use of Artificial Neural Network- Hidden Markov Model (ANN/HMM) hybrid systems for Automatic Speech Recognition (ASR). In this paper we report results of a DBN-pretrained context-dependent ANN/HMM system trained on two datasets that are much larger than any reported previously with DBN-pretrained ANN/HMM

State of the art speech recognition systems rely on preprocessed speech features such as Mel cepstrum or linear predictive coding coefficients that collapse high dimensional speech sound waves into low dimensional encodings. While these have been successfully applied in speech recognition systems, such low dimensional encodings may lose some relevant information and express other information in a way that makes it difficult to use for discrimination. Higher dimensional encodings could both improve performance in recognition tasks, and also be applied to speech synthesis by better modeling the statistical structure of the sound waves. In this paper we present a novel approach for modeling speech sound waves using a Restricted Boltzmann machine (RBM) with a novel type of hidden variable and we report initial results demonstrating phoneme recognition performance better than the current state-of-the-art for methods based on Mel cepstrum coefficients.

MOTIVATION: Electron cryo-microscopy can be used to infer 3D structures of large macromolecules with high resolution, but the large amounts of data captured necessitate the development of appropriate statistical models to describe the data generation process, and to perform structure inference. We present a new method for performing ab initio inference of the 3D structures of macromolecules from single particle electron cryo-microscopy experiments using class average images. RESULTS: We demonstrate this algorithm on one phantom, one synthetic dataset and three real (experimental) datasets (ATP synthase, V-type ATPase and GroEL). Structures consistent with the known structures were inferred for all datasets. AVAILABILITY: The software and source code for this method is available for download from our website: http://compbio.cs.toronto.edu/cryoem/.

Independent, greedy collection of data events using simple heuristics results in massive over-sampling of the prominent data features in large-scale studies over what should be achievable through ¿intelligent", online acquisition of such data. As a result, data generated are more aptly described as a collection of a large number of small experiments rather than a true large-scale experiment. Nevertheless, achieving ¿intelligent¿, online control requires tight interplay between state-of-the-art, data-intensive computing infrastructure developments and analytical algorithms. In this paper, we propose a Software Architecture for Mass spectrometry-based Proteomics coupled with Liquid chromatography Experiments (SAMPLE) to develop an ¿intelligent¿ online control and analysis system to significantly enhance the information content from each sensor (in this case, a mass spectrometer). Using online analysis of data events as they are collected and decision theory to optimize the collection of events during an experiment, we aim to maximize the information content generated during an experiment by the use of pre-existing knowledge to optimize the dynamic collection of events.

BACKGROUND: Data generated from liquid chromatography coupled to high-resolution mass spectrometry (LC-MS)-based studies of a biological sample can contain large amounts of biologically significant information in the form of proteins, peptides, and metabolites. Interpreting this data involves inferring the masses and abundances of biomolecules injected into the instrument. Because of the inherent complexity of mass spectral patterns produced by these biomolecules, the analysis is significantly enhanced by using visualization capabilities to inspect and confirm results. In this paper we describe Decon2LS, an open-source software package for automated processing and visualization of high-resolution MS data. Drawing extensively on algorithms developed over the last ten years for ICR2LS, Decon2LS packages the algorithms as a rich set of modular, reusable processing classes for performing diverse functions such as reading raw data, routine peak finding, theoretical isotope distribution modelling, and deisotoping. Because the source code is openly available, these functionalities can now be used to build derivative applications in relatively fast manner. In addition, Decon2LS provides an extensive set of visualization tools, such as high performance chart controls. RESULTS: With a variety of options that include peak processing, deisotoping, isotope composition, etc, Decon2LS supports processing of multiple raw data formats. Deisotoping can be performed on an individual scan, an individual dataset, or on multiple datasets using batch processing. Other processing options include creating a two dimensional view of mass and liquid chromatography (LC) elution time features, generating spectrum files for tandem MS data, creating total intensity chromatograms, and visualizing theoretical peptide profiles. Application of Decon2LS to deisotope different datasets obtained across different instruments yielded a high number of features that can be used to identify and quantify peptides in the biological sample. CONCLUSION: Decon2LS is an efficient software package for discovering and visualizing features in proteomics studies that require automated interpretation of mass spectra. Besides being easy to use, fast, and reliable, Decon2LS is also open-source, which allows developers in the proteomics and bioinformatics communities to reuse and refine the algorithms to meet individual needs.Decon2LS source code, installer, and tutorials may be downloaded free of charge at http://http:/ncrr.pnl.gov/software/.

UNLABELLED: DeconMSn accurately determines the monoisotopic mass and charge state of parent ions from high-resolution tandem mass spectrometry data, offering significant improvement for LTQ_FT and LTQ_Orbitrap instruments over the commercially delivered Thermo Fisher Scientific's extract_msn tool. Optimal parent ion mass tolerance values can be determined using accurate mass information, thus improving peptide identifications for high-mass measurement accuracy experiments. For low-resolution data from LCQ and LTQ instruments, DeconMSn incorporates a support-vector-machine-based charge detection algorithm that identifies the most likely charge of a parent species through peak characteristics of its fragmentation pattern. AVAILABILITY: http://ncrr.pnl.gov/software/ or http://www.proteomicsresource.org/.

UNLABELLED: Data Analysis Tool Extension (DAnTE) is a statistical tool designed to address challenges associated with quantitative bottom-up, shotgun proteomics data. This tool has also been demonstrated for microarray data and can easily be extended to other high-throughput data types. DAnTE features selected normalization methods, missing value imputation algorithms, peptide-to-protein rollup methods, an extensive array of plotting functions and a comprehensive hypothesis-testing scheme that can handle unbalanced data and random effects. The graphical user interface (GUI) is designed to be very intuitive and user friendly. AVAILABILITY: DAnTE may be downloaded free of charge at http://omics.pnl.gov/software/. SUPPLEMENTARY INFORMATION: An example dataset with instructions on how to perform a series of analysis steps is available at http://omics.pnl.gov/software/

Abstract Summary: The accurate mass and time (AMT) tag approach is used for analysis of large scale experiments by combining information generated over multiple datasets and instrument types. The VIPER software package is one of the key components of the data processing pipeline and implements automated algorithms to discover LC-MS features, align and match these LC-MS features to a database of peptides previously identified in LC-MS/MS analyses, and identify and quantify pairs of isotopically labeled peptides. Availability: VIPER may be downloaded free of charge at http://ncrr.pnl.gov/software/ Contact: [email protected] or [email protected] Supplementary information: Supplementary data are available at Bioinformatics online.