Audio-Visual Approach For Multimodal Concurrent Speaker Detection

Concurrent Speaker Detection (CSD), the task of identifying the presence and overlap of active speakers in an audio signal, is crucial for many audio tasks such as meeting transcription, speaker diarization, and speech separation. This study introduces a multimodal deep learning approach that leverages both audio and visual information. The proposed model employs an early fusion strategy combining audio and visual features through cross-modal attention mechanisms, with a learnable [CLS] token capturing the relevant audio-visual relationships. The model is extensively evaluated on two real-world datasets, AMI and the recently introduced EasyCom dataset. Experiments validate the effectiveness of the multimodal fusion strategy. Ablation studies further support the design choices and the training procedure of the model. As this is the first work reporting CSD results on the challenging EasyCom dataset, the findings demonstrate the potential of the proposed multimodal approach for CSD in real-world scenarios.

peerRTF: Robust MVDR Beamforming Using Graph Convolutional Network

Accurate and reliable identification of the RTF between microphones with respect to a desired source is an essential component in the design of microphone array beamformers, specifically the MVDR criterion. Since an accurate estimation of the RTF in a noisy and reverberant environment is a cumbersome task, we aim at leveraging prior knowledge of the acoustic enclosure to robustify the RTF estimation by learning the RTF manifold. In this paper, we present a novel robust RTF identification method, tested and trained with real recordings, which relies on learning the RTF manifold using a GCN to infer a robust representation of the RTF in a confined area, and consequently enhance the beamformer's performance.

Multi-Microphone Speech Emotion Recognition using the Hierarchical Token-semantic Audio Transformer Architecture

Most emotion recognition systems fail in real-life situations (in the wild scenarios) where the audio is contaminated by reverberation. Our study explores new methods to alleviate the performance degradation of Speech Emotion Recognition (SER) algorithms and develop a more robust system for adverse conditions. We propose processing multi-microphone signals to address these challenges and improve emotion classification accuracy. We adopt a state-of-the-art transformer model, the Hierarchical Token-semantic Audio Transformer (HTS-AT), to handle multi-channel audio inputs. We evaluate two strategies: averaging mel-spectrograms across channels and summing patch-embedded representations. Our multimicrophone model achieves superior performance compared to single-channel baselines when tested on real-world reverberant environments.

RevRIR: Joint Reverberant Speech and Room Impulse Response Embedding using Contrastive Learning with Application to Room Shape Classification

This paper focuses on room fingerprinting, a task involving the analysis of an audio recording to determine the specific volume and shape of the room in which it was captured. While it is relatively straightforward to determine the basic room parameters from the Room Impulse Responses (RIR), doing so from a speech signal is a cumbersome task. To address this challenge, we introduce a dual-encoder architecture that facilitates the estimation of room parameters directly from speech utterances. During pre-training, one encoder receives the RIR while the other processes the reverberant speech signal. A contrastive loss function is employed to embed the speech and the acoustic response jointly. In the fine-tuning stage, the specific classification task is trained. In the test phase, only the reverberant utterance is available, and its embedding is used for the task of room shape classification. The proposed scheme is extensively evaluated using simulated acoustic environments.

SingIt! Singer Voice Transformation

In this paper, we propose a model which can generate a singing voice from normal speech utterance by harnessing zero-shot, many-to-many style transfer learning. Our goal is to give anyone the opportunity to sing any song in a timely manner. We present a system comprising several available blocks, as well as a modified auto-encoder, and show how this highly-complex challenge can be achieved by tailoring rather simple solutions together. We demonstrate the applicability of the proposed system using a group of 25 non-expert listeners. Samples of the data generated from our model are provided.

Socially Pertinent Robots in Gerontological Healthcare

Despite the many recent achievements in developing and deploying social robotics, there are still many underexplored environments and applications for which systematic evaluation of such systems by end-users is necessary. While several robotic platforms have been used in gerontological healthcare, the question of whether or not a social interactive robot with multi-modal conversational capabilities will be useful and accepted in real-life facilities is yet to be answered. This paper is an attempt to partially answer this question, via two waves of experiments with patients and companions in a day-care gerontological facility in Paris with a full-sized humanoid robot endowed with social and conversational interaction capabilities. The software architecture, developed during the H2020 SPRING project, together with the experimental protocol, allowed us to evaluate the acceptability (AES) and usability (SUS) with more than 60 end-users. Overall, the users are receptive to this technology, especially when the robot perception and action skills are robust to environmental clutter and flexible to handle a plethora of different interactions.

Concurrent Speaker Detection: A multi-microphone Transformer-Based Approach

We present a deep-learning approach for the task of Concurrent Speaker Detection (CSD) using a modified transformer model. Our model is designed to handle multi-microphone data but can also work in the single-microphone case. The method can classify audio segments into one of three classes: 1) no speech activity (noise only), 2) only a single speaker is active, and 3) more than one speaker is active. We incorporate a Cost-Sensitive (CS) loss and a confidence calibration to the training procedure. The approach is evaluated using three real-world databases: AMI, AliMeeting, and CHiME 5, demonstrating an improvement over existing approaches.

Comparison of Frequency-Fusion Mechanisms for Binaural Direction-of-Arrival Estimation for Multiple Speakers

To estimate the direction of arrival (DOA) of multiple speakers with methods that use prototype transfer functions, frequency-dependent spatial spectra (SPS) are usually constructed. To make the DOA estimation robust, SPS from different frequencies can be combined. According to how the SPS are combined, frequency fusion mechanisms are categorized into narrowband, broadband, or speaker-grouped, where the latter mechanism requires a speaker-wise grouping of frequencies. For a binaural hearing aid setup, in this paper we propose an interaural time difference (ITD)-based speaker-grouped frequency fusion mechanism. By exploiting the DOA dependence of ITDs, frequencies can be grouped according to a common ITD and be used for DOA estimation of the respective speaker. We apply the proposed ITD-based speaker-grouped frequency fusion mechanism for different DOA estimation methods, namely the multiple signal classification, steered response power and a recently published method based on relative transfer function (RTF) vectors. In our experiments, we compare DOA estimation with different fusion mechanisms. For all considered DOA estimation methods, the proposed ITD-based speaker-grouped frequency fusion mechanism results in a higher DOA estimation accuracy compared with the narrowband and broadband fusion mechanisms.

Single-Microphone Speaker Separation and Voice Activity Detection in Noisy and Reverberant Environments

Speech separation involves extracting an individual speaker's voice from a multi-speaker audio signal. The increasing complexity of real-world environments, where multiple speakers might converse simultaneously, underscores the importance of effective speech separation techniques. This work presents a single-microphone speaker separation network with TF attention aiming at noisy and reverberant environments. We dub this new architecture as Separation TF Attention Network (Sep-TFAnet). In addition, we present a variant of the separation network, dubbed $ \text{Sep-TFAnet}^{\text{VAD}}$, which incorporates a voice activity detector (VAD) into the separation network. The separation module is based on a temporal convolutional network (TCN) backbone inspired by the Conv-Tasnet architecture with multiple modifications. Rather than a learned encoder and decoder, we use short-time Fourier transform (STFT) and inverse short-time Fourier transform (iSTFT) for the analysis and synthesis, respectively. Our system is specially developed for human-robotic interactions and should support online mode. The separation capabilities of $ \text{Sep-TFAnet}^{\text{VAD}}$ and Sep-TFAnet were evaluated and extensively analyzed under several acoustic conditions, demonstrating their advantages over competing methods. Since separation networks trained on simulated data tend to perform poorly on real recordings, we also demonstrate the ability of the proposed scheme to better generalize to realistic examples recorded in our acoustic lab by a humanoid robot. Project page: https://Sep-TFAnet.github.io

LipVoicer: Generating Speech from Silent Videos Guided by Lip Reading

Lip-to-speech involves generating a natural-sounding speech synchronized with a soundless video of a person talking. Despite recent advances, current methods still cannot produce high-quality speech with high levels of intelligibility for challenging and realistic datasets such as LRS3. In this work, we present LipVoicer, a novel method that generates high-quality speech, even for in-the-wild and rich datasets, by incorporating the text modality. Given a silent video, we first predict the spoken text using a pre-trained lip-reading network. We then condition a diffusion model on the video and use the extracted text through a classifier-guidance mechanism where a pre-trained ASR serves as the classifier. LipVoicer outperforms multiple lip-to-speech baselines on LRS2 and LRS3, which are in-the-wild datasets with hundreds of unique speakers in their test set and an unrestricted vocabulary. Moreover, our experiments show that the inclusion of the text modality plays a major role in the intelligibility of the produced speech, readily perceptible while listening, and is empirically reflected in the substantial reduction of the WER metric. We demonstrate the effectiveness of LipVoicer through human evaluation, which shows that it produces more natural and synchronized speech signals compared to competing methods. Finally, we created a demo showcasing LipVoicer's superiority in producing natural, synchronized, and intelligible speech, providing additional evidence of its effectiveness. Project page: https://lipvoicer.github.io