Daniel Stoller

Josh Gardner, Simon Durand, Daniel Stoller et al.

Music has a unique and complex structure which is challenging for both expert humans and existing AI systems to understand, and presents unique challenges relative to other forms of audio. We present LLark, an instruction-tuned multimodal model for \emph{music} understanding. We detail our process for dataset creation, which involves augmenting the annotations of diverse open-source music datasets and converting them to a unified instruction-tuning format. We propose a multimodal architecture for LLark, integrating a pretrained generative model for music with a pretrained language model. In evaluations on three types of tasks (music understanding, captioning, reasoning), we show that LLark matches or outperforms existing baselines in music understanding, and that humans show a high degree of agreement with its responses in captioning and reasoning tasks. LLark is trained entirely from open-source music data and models, and we make our training code available along with the release of this paper. Additional results and audio examples are at https://bit.ly/llark, and our source code is available at https://github.com/spotify-research/llark .

Simon Durand, Daniel Stoller, Sebastian Ewert

Lyrics alignment gained considerable attention in recent years. State-of-the-art systems either re-use established speech recognition toolkits, or design end-to-end solutions involving a Connectionist Temporal Classification (CTC) loss. However, both approaches suffer from specific weaknesses: toolkits are known for their complexity, and CTC systems use a loss designed for transcription which can limit alignment accuracy. In this paper, we use instead a contrastive learning procedure that derives cross-modal embeddings linking the audio and text domains. This way, we obtain a novel system that is simple to train end-to-end, can make use of weakly annotated training data, jointly learns a powerful text model, and is tailored to alignment. The system is not only the first to yield an average absolute error below 0.2 seconds on the standard Jamendo dataset but it is also robust to other languages, even when trained on English data only. Finally, we release word-level alignments for the JamendoLyrics Multi-Lang dataset.

Daniel Stoller, Mi Tian, Sebastian Ewert et al.

Convolutional neural networks (CNNs) with dilated filters such as the Wavenet or the Temporal Convolutional Network (TCN) have shown good results in a variety of sequence modelling tasks. However, efficiently modelling long-term dependencies in these sequences is still challenging. Although the receptive field of these models grows exponentially with the number of layers, computing the convolutions over very long sequences of features in each layer is time and memory-intensive, prohibiting the use of longer receptive fields in practice. To increase efficiency, we make use of the "slow feature" hypothesis stating that many features of interest are slowly varying over time. For this, we use a U-Net architecture that computes features at multiple time-scales and adapt it to our auto-regressive scenario by making convolutions causal. We apply our model ("Seq-U-Net") to a variety of tasks including language and audio generation. In comparison to TCN and Wavenet, our network consistently saves memory and computation time, with speed-ups for training and inference of over 4x in the audio generation experiment in particular, while achieving a comparable performance in all tasks.

Daniel Stoller, Sebastian Ewert, Simon Dixon

Generative adversarial networks (GANs) have shown great success in applications such as image generation and inpainting. However, they typically require large datasets, which are often not available, especially in the context of prediction tasks such as image segmentation that require labels. Therefore, methods such as the CycleGAN use more easily available unlabelled data, but do not offer a way to leverage additional labelled data for improved performance. To address this shortcoming, we show how to factorise the joint data distribution into a set of lower-dimensional distributions along with their dependencies. This allows splitting the discriminator in a GAN into multiple "sub-discriminators" that can be independently trained from incomplete observations. Their outputs can be combined to estimate the density ratio between the joint real and the generator distribution, which enables training generators as in the original GAN framework. We apply our method to image generation, image segmentation and audio source separation, and obtain improved performance over a standard GAN when additional incomplete training examples are available. For the Cityscapes segmentation task in particular, our method also improves accuracy by an absolute 14.9% over CycleGAN while using only 25 additional paired examples.

Saumitra Mishra, Daniel Stoller, Emmanouil Benetos et al.

One way to interpret trained deep neural networks (DNNs) is by inspecting characteristics that neurons in the model respond to, such as by iteratively optimising the model input (e.g., an image) to maximally activate specific neurons. However, this requires a careful selection of hyper-parameters to generate interpretable examples for each neuron of interest, and current methods rely on a manual, qualitative evaluation of each setting, which is prohibitively slow. We introduce a new metric that uses Fréchet Inception Distance (FID) to encourage similarity between model activations for real and generated data. This provides an efficient way to evaluate a set of generated examples for each setting of hyper-parameters. We also propose a novel GAN-based method for generating explanations that enables an efficient search through the input space and imposes a strong prior favouring realistic outputs. We apply our approach to a classification model trained to predict whether a music audio recording contains singing voice. Our results suggest that this proposed metric successfully selects hyper-parameters leading to interpretable examples, avoiding the need for manual evaluation. Moreover, we see that examples synthesised to maximise or minimise the predicted probability of singing voice presence exhibit vocal or non-vocal characteristics, respectively, suggesting that our approach is able to generate suitable explanations for understanding concepts learned by a neural network.

Bhusan Chettri, Daniel Stoller, Veronica Morfi et al.

Detecting spoofing attempts of automatic speaker verification (ASV) systems is challenging, especially when using only one modeling approach. For robustness, we use both deep neural networks and traditional machine learning models and combine them as ensemble models through logistic regression. They are trained to detect logical access (LA) and physical access (PA) attacks on the dataset released as part of the ASV Spoofing and Countermeasures Challenge 2019. We propose dataset partitions that ensure different attack types are present during training and validation to improve system robustness. Our ensemble model outperforms all our single models and the baselines from the challenge for both attack types. We investigate why some models on the PA dataset strongly outperform others and find that spoofed recordings in the dataset tend to have longer silences at the end than genuine ones. By removing them, the PA task becomes much more challenging, with the tandem detection cost function (t-DCF) of our best single model rising from 0.1672 to 0.5018 and equal error rate (EER) increasing from 5.98% to 19.8% on the development set.

Daniel Stoller, Simon Durand, Sebastian Ewert

Time-aligned lyrics can enrich the music listening experience by enabling karaoke, text-based song retrieval and intra-song navigation, and other applications. Compared to text-to-speech alignment, lyrics alignment remains highly challenging, despite many attempts to combine numerous sub-modules including vocal separation and detection in an effort to break down the problem. Furthermore, training required fine-grained annotations to be available in some form. Here, we present a novel system based on a modified Wave-U-Net architecture, which predicts character probabilities directly from raw audio using learnt multi-scale representations of the various signal components. There are no sub-modules whose interdependencies need to be optimized. Our training procedure is designed to work with weak, line-level annotations available in the real world. With a mean alignment error of 0.35s on a standard dataset our system outperforms the state-of-the-art by an order of magnitude.

Daniel Stoller, Sebastian Ewert, Simon Dixon

Models for audio source separation usually operate on the magnitude spectrum, which ignores phase information and makes separation performance dependant on hyper-parameters for the spectral front-end. Therefore, we investigate end-to-end source separation in the time-domain, which allows modelling phase information and avoids fixed spectral transformations. Due to high sampling rates for audio, employing a long temporal input context on the sample level is difficult, but required for high quality separation results because of long-range temporal correlations. In this context, we propose the Wave-U-Net, an adaptation of the U-Net to the one-dimensional time domain, which repeatedly resamples feature maps to compute and combine features at different time scales. We introduce further architectural improvements, including an output layer that enforces source additivity, an upsampling technique and a context-aware prediction framework to reduce output artifacts. Experiments for singing voice separation indicate that our architecture yields a performance comparable to a state-of-the-art spectrogram-based U-Net architecture, given the same data. Finally, we reveal a problem with outliers in the currently used SDR evaluation metrics and suggest reporting rank-based statistics to alleviate this problem.

Daniel Stoller, Sebastian Ewert, Simon Dixon

A main challenge in applying deep learning to music processing is the availability of training data. One potential solution is Multi-task Learning, in which the model also learns to solve related auxiliary tasks on additional datasets to exploit their correlation. While intuitive in principle, it can be challenging to identify related tasks and construct the model to optimally share information between tasks. In this paper, we explore vocal activity detection as an additional task to stabilise and improve the performance of vocal separation. Further, we identify problematic biases specific to each dataset that could limit the generalisation capability of separation and detection models, to which our proposed approach is robust. Experiments show improved performance in separation as well as vocal detection compared to single-task baselines. However, we find that the commonly used Signal-to-Distortion Ratio (SDR) metrics did not capture the improvement on non-vocal sections, indicating the need for improved evaluation methodologies.

Daniel Stoller, Sebastian Ewert, Simon Dixon

The state of the art in music source separation employs neural networks trained in a supervised fashion on multi-track databases to estimate the sources from a given mixture. With only few datasets available, often extensive data augmentation is used to combat overfitting. Mixing random tracks, however, can even reduce separation performance as instruments in real music are strongly correlated. The key concept in our approach is that source estimates of an optimal separator should be indistinguishable from real source signals. Based on this idea, we drive the separator towards outputs deemed as realistic by discriminator networks that are trained to tell apart real from separator samples. This way, we can also use unpaired source and mixture recordings without the drawbacks of creating unrealistic music mixtures. Our framework is widely applicable as it does not assume a specific network architecture or number of sources. To our knowledge, this is the first adoption of adversarial training for music source separation. In a prototype experiment for singing voice separation, separation performance increases with our approach compared to purely supervised training.