Abstract:Discovering a lexicon from unlabeled audio is a longstanding challenge for zero-resource speech processing. One approach is to search for frequently occurring patterns in speech. We revisit this idea with DUSTED: Discrete Unit Spoken-TErm Discovery. Leveraging self-supervised models, we encode input audio into sequences of discrete units. Next, we find repeated patterns by searching for similar unit sub-sequences, inspired by alignment algorithms from bioinformatics. Since discretization discards speaker information, DUSTED finds better matches across speakers, improving the coverage and consistency of the discovered patterns. We demonstrate these improvements on the ZeroSpeech Challenge, achieving state-of-the-art results on the spoken-term discovery track. Finally, we analyze the duration distribution of the patterns, showing that our method finds longer word- or phrase-like terms.
Abstract:Real world deployments of word alignment are almost certain to cover both high and low resource languages. However, the state-of-the-art for this task recommends a different model class depending on the availability of gold alignment training data for a particular language pair. We propose BinaryAlign, a novel word alignment technique based on binary sequence labeling that outperforms existing approaches in both scenarios, offering a unifying approach to the task. Additionally, we vary the specific choice of multilingual foundation model, perform stratified error analysis over alignment error type, and explore the performance of BinaryAlign on non-English language pairs. We make our source code publicly available.
Abstract:Grammatical Error Detection (GED) methods rely heavily on human annotated error corpora. However, these annotations are unavailable in many low-resource languages. In this paper, we investigate GED in this context. Leveraging the zero-shot cross-lingual transfer capabilities of multilingual pre-trained language models, we train a model using data from a diverse set of languages to generate synthetic errors in other languages. These synthetic error corpora are then used to train a GED model. Specifically we propose a two-stage fine-tuning pipeline where the GED model is first fine-tuned on multilingual synthetic data from target languages followed by fine-tuning on human-annotated GED corpora from source languages. This approach outperforms current state-of-the-art annotation-free GED methods. We also analyse the errors produced by our method and other strong baselines, finding that our approach produces errors that are more diverse and more similar to human errors.
Abstract:We introduce UPose3D, a novel approach for multi-view 3D human pose estimation, addressing challenges in accuracy and scalability. Our method advances existing pose estimation frameworks by improving robustness and flexibility without requiring direct 3D annotations. At the core of our method, a pose compiler module refines predictions from a 2D keypoints estimator that operates on a single image by leveraging temporal and cross-view information. Our novel cross-view fusion strategy is scalable to any number of cameras, while our synthetic data generation strategy ensures generalization across diverse actors, scenes, and viewpoints. Finally, UPose3D leverages the prediction uncertainty of both the 2D keypoint estimator and the pose compiler module. This provides robustness to outliers and noisy data, resulting in state-of-the-art performance in out-of-distribution settings. In addition, for in-distribution settings, UPose3D yields a performance rivaling methods that rely on 3D annotated data, while being the state-of-the-art among methods relying only on 2D supervision.
Abstract:Audio diffusion models can synthesize a wide variety of sounds. Existing models often operate on the latent domain with cascaded phase recovery modules to reconstruct waveform. This poses challenges when generating high-fidelity audio. In this paper, we propose EDMSound, a diffusion-based generative model in spectrogram domain under the framework of elucidated diffusion models (EDM). Combining with efficient deterministic sampler, we achieved similar Fr\'echet audio distance (FAD) score as top-ranked baseline with only 10 steps and reached state-of-the-art performance with 50 steps on the DCASE2023 foley sound generation benchmark. We also revealed a potential concern regarding diffusion based audio generation models that they tend to generate samples with high perceptual similarity to the data from training data. Project page: https://agentcooper2002.github.io/EDMSound/
Abstract:Voice conversion aims to transform source speech into a different target voice. However, typical voice conversion systems do not account for rhythm, which is an important factor in the perception of speaker identity. To bridge this gap, we introduce Urhythmic-an unsupervised method for rhythm conversion that does not require parallel data or text transcriptions. Using self-supervised representations, we first divide source audio into segments approximating sonorants, obstruents, and silences. Then we model rhythm by estimating speaking rate or the duration distribution of each segment type. Finally, we match the target speaking rate or rhythm by time-stretching the speech segments. Experiments show that Urhythmic outperforms existing unsupervised methods in terms of quality and prosody. Code and checkpoints: https://github.com/bshall/urhythmic. Audio demo page: https://ubisoft-laforge.github.io/speech/urhythmic.
Abstract:We present ZeroEGGS, a neural network framework for speech-driven gesture generation with zero-shot style control by example. This means style can be controlled via only a short example motion clip, even for motion styles unseen during training. Our model uses a Variational framework to learn a style embedding, making it easy to modify style through latent space manipulation or blending and scaling of style embeddings. The probabilistic nature of our framework further enables the generation of a variety of outputs given the same input, addressing the stochastic nature of gesture motion. In a series of experiments, we first demonstrate the flexibility and generalizability of our model to new speakers and styles. In a user study, we then show that our model outperforms previous state-of-the-art techniques in naturalness of motion, appropriateness for speech, and style portrayal. Finally, we release a high-quality dataset of full-body gesture motion including fingers, with speech, spanning across 19 different styles.
Abstract:The goal of voice conversion is to transform source speech into a target voice, keeping the content unchanged. In this paper, we focus on self-supervised representation learning for voice conversion. Specifically, we compare discrete and soft speech units as input features. We find that discrete representations effectively remove speaker information but discard some linguistic content - leading to mispronunciations. As a solution, we propose soft speech units. To learn soft units, we predict a distribution over discrete speech units. By modeling uncertainty, soft units capture more content information, improving the intelligibility and naturalness of converted speech. Samples available at https://ubisoft-laforge.github.io/speech/soft-vc/
Abstract:This paper presents Daft-Exprt, a multi-speaker acoustic model advancing the state-of-the-art on inter-speaker and inter-text prosody transfer. This improvement is achieved using FiLM conditioning layers, alongside adversarial training that encourages disentanglement between prosodic information and speaker identity. The acoustic model inherits attractive qualities from FastSpeech 2, such as fast inference and local prosody attributes prediction for finer grained control over generation. Experimental results show that Daft-Exprt significantly outperforms strong baselines on prosody transfer tasks, while yielding naturalness comparable to state-of-the-art expressive models. Moreover, results indicate that adversarial training effectively discards speaker identity information from the prosody representation, which ensures Daft-Exprt will consistently generate speech with the desired voice. We publicly release our code and provide speech samples from our experiments.
Abstract:Non-intrusive load monitoring (NILM) is a technique that uses a single sensor to measure the total power consumption of a building. Using an energy disaggregation method, the consumption of individual appliances can be estimated from the aggregate measurement. Recent disaggregation algorithms have significantly improved the performance of NILM systems. However, the generalization capability of these methods to different houses as well as the disaggregation of multi-state appliances are still major challenges. In this paper we address these issues and propose an energy disaggregation approach based on the variational autoencoders (VAE) framework. The probabilistic encoder makes this approach an efficient model for encoding information relevant to the reconstruction of the target appliance consumption. In particular, the proposed model accurately generates more complex load profiles, thus improving the power signal reconstruction of multi-state appliances. Moreover, its regularized latent space improves the generalization capabilities of the model across different houses. The proposed model is compared to state-of-the-art NILM approaches on the UK-DALE dataset, and yields competitive results. The mean absolute error reduces by 18% on average across all appliances compared to the state-of-the-art. The F1-Score increases by more than 11%, showing improvements for the detection of the target appliance in the aggregate measurement.