While text-based event extraction has been an active research area and has seen successful application in many domains, extracting semantic events from speech directly is an under-explored problem. In this paper, we introduce the Speech Event Extraction (SpeechEE) task and construct three synthetic training sets and one human-spoken test set. Compared to event extraction from text, SpeechEE poses greater challenges mainly due to complex speech signals that are continuous and have no word boundaries. Additionally, unlike perceptible sound events, semantic events are more subtle and require a deeper understanding. To tackle these challenges, we introduce a sequence-to-structure generation paradigm that can produce events from speech signals in an end-to-end manner, together with a conditioned generation method that utilizes speech recognition transcripts as the contextual clue. We further propose to represent events with a flat format to make outputs more natural language-like. Our experimental results show that our method brings significant improvements on all datasets, achieving a maximum F1 gain of 10.7%. The code and datasets are released on https://github.com/jodie-kang/SpeechEE.
Large language models (LLM) have demonstrated their abilities to solve various natural language processing tasks through dialogue-based interactions. For instance, research indicates that LLMs can achieve competitive performance in offline machine translation tasks for high-resource languages. However, applying LLMs to simultaneous machine translation (SimulMT) poses many challenges, including issues related to the training-inference mismatch arising from different decoding patterns. In this paper, we explore the feasibility of utilizing LLMs for SimulMT. Building upon conventional approaches, we introduce a simple yet effective mixture policy that enables LLMs to engage in SimulMT without requiring additional training. Furthermore, after Supervised Fine-Tuning (SFT) on a mixture of full and prefix sentences, the model exhibits significant performance improvements. Our experiments, conducted with Llama2-7B-chat on nine language pairs from the MUST-C dataset, demonstrate that LLM can achieve translation quality and latency comparable to dedicated SimulMT models.
Pre-trained speech encoders have been central to pushing state-of-the-art results across various speech understanding and generation tasks. Nonetheless, the capabilities of these encoders in low-resource settings are yet to be thoroughly explored. To address this, we conduct a comprehensive set of experiments using a representative set of 3 state-of-the-art encoders (Wav2vec2, WavLM, Whisper) in the low-resource setting across 7 speech understanding and generation tasks. We provide various quantitative and qualitative analyses on task performance, convergence speed, and representational properties of the encoders. We observe a connection between the pre-training protocols of these encoders and the way in which they capture information in their internal layers. In particular, we observe the Whisper encoder exhibits the greatest low-resource capabilities on content-driven tasks in terms of performance and convergence speed.
It remains a question that how simultaneous interpretation (SI) data affects simultaneous machine translation (SiMT). Research has been limited due to the lack of a large-scale training corpus. In this work, we aim to fill in the gap by introducing NAIST-SIC-Aligned, which is an automatically-aligned parallel English-Japanese SI dataset. Starting with a non-aligned corpus NAIST-SIC, we propose a two-stage alignment approach to make the corpus parallel and thus suitable for model training. The first stage is coarse alignment where we perform a many-to-many mapping between source and target sentences, and the second stage is fine-grained alignment where we perform intra- and inter-sentence filtering to improve the quality of aligned pairs. To ensure the quality of the corpus, each step has been validated either quantitatively or qualitatively. This is the first open-sourced large-scale parallel SI dataset in the literature. We also manually curated a small test set for evaluation purposes. We hope our work advances research on SI corpora construction and SiMT. Please find our data at \url{https://github.com/mingzi151/AHC-SI}.
Over the past few decades, multimodal emotion recognition has made remarkable progress with the development of deep learning. However, existing technologies are difficult to meet the demand for practical applications. To improve the robustness, we launch a Multimodal Emotion Recognition Challenge (MER 2023) to motivate global researchers to build innovative technologies that can further accelerate and foster research. For this year's challenge, we present three distinct sub-challenges: (1) MER-MULTI, in which participants recognize both discrete and dimensional emotions; (2) MER-NOISE, in which noise is added to test videos for modality robustness evaluation; (3) MER-SEMI, which provides large amounts of unlabeled samples for semi-supervised learning. In this paper, we test a variety of multimodal features and provide a competitive baseline for each sub-challenge. Our system achieves 77.57% on the F1 score and 0.82 on the mean squared error (MSE) for MER-MULTI, 69.82% on the F1 score and 1.12 on MSE for MER-NOISE, and 86.75% on the F1 score for MER-SEMI, respectively. Baseline code is available at https://github.com/zeroQiaoba/MER2023-Baseline.
Multimodal emotion recognition leverages complementary information across modalities to gain performance. However, we cannot guarantee that the data of all modalities are always present in practice. In the studies to predict the missing data across modalities, the inherent difference between heterogeneous modalities, namely the modality gap, presents a challenge. To address this, we propose to use invariant features for a missing modality imagination network (IF-MMIN) which includes two novel mechanisms: 1) an invariant feature learning strategy that is based on the central moment discrepancy (CMD) distance under the full-modality scenario; 2) an invariant feature based imagination module (IF-IM) to alleviate the modality gap during the missing modalities prediction, thus improving the robustness of multimodal joint representation. Comprehensive experiments on the benchmark dataset IEMOCAP demonstrate that the proposed model outperforms all baselines and invariantly improves the overall emotion recognition performance under uncertain missing-modality conditions. We release the code at: https://github.com/ZhuoYulang/IF-MMIN.
Pre-trained speech Transformers have facilitated great success across various speech processing tasks. However, fine-tuning these encoders for downstream tasks require sufficiently large training data to converge or to achieve state-of-the-art. In text domain this has been partly attributed to sub-optimality of the representation space in pre-trained Transformers. In this work, we take a sober look into pre-trained speech encoders and rewire their representation space without requiring any task-specific labels. Our method utilises neutrally synthesised version of audio inputs along with frame masking to construct positive pairs for contrastive self-supervised learning. When used for augmenting the wav2vec 2 encoder, we observe consistent improvement of isotropy in the representation space. Our experiments on 6 speech processing tasks, exhibit a significant convergence speedup during task fine-tuning as well as consistent task improvement, specially in low-resource settings.
Relation extraction typically aims to extract semantic relationships between entities from the unstructured text. One of the most essential data sources for relation extraction is the spoken language, such as interviews and dialogues. However, the error propagation introduced in automatic speech recognition (ASR) has been ignored in relation extraction, and the end-to-end speech-based relation extraction method has been rarely explored. In this paper, we propose a new listening information extraction task, i.e., speech relation extraction. We construct the training dataset for speech relation extraction via text-to-speech systems, and we construct the testing dataset via crowd-sourcing with native English speakers. We explore speech relation extraction via two approaches: the pipeline approach conducting text-based extraction with a pretrained ASR module, and the end2end approach via a new proposed encoder-decoder model, or what we called SpeechRE. We conduct comprehensive experiments to distinguish the challenges in speech relation extraction, which may shed light on future explorations. We share the code and data on https://github.com/wutong8023/SpeechRE.
Pre-trained speech Transformers in speech translation (ST) have facilitated state-of-the-art (SotA) results; yet, using such encoders is computationally expensive. To improve this, we present a novel Reducer Adaptor block, RedApt, that could be seamlessly integrated within any Transformer-based speech encoding architecture. Integrating the pretrained wav2vec 2 speech encoder with RedAptbrings 41% speedup, 33% memory reduction with 24% fewer FLOPs at inference. To our positive surprise, our ST model with RedApt outperforms the SotA architecture by an average of 0.68 BLEU score on 8 language pairs from Must-C.
Training end-to-end speech translation (ST) systems requires sufficiently large-scale data, which is unavailable for most language pairs and domains. One practical solution to the data scarcity issue is to convert machine translation data (MT) to ST data via text-to-speech (TTS) systems. Yet, using TTS systems can be tedious and slow, as the conversion needs to be done for each MT dataset. In this work, we propose a simple, scalable and effective data augmentation technique, i.e., SpokenVocab, to convert MT data to ST data on-the-fly. The idea is to retrieve and stitch audio snippets from a SpokenVocab bank according to words in an MT sequence. Our experiments on multiple language pairs from Must-C show that this method outperforms strong baselines by an average of 1.83 BLEU scores, and it performs equally well as TTS-generated speech. We also showcase how SpokenVocab can be applied in code-switching ST for which often no TTS systems exit. Our code is available at https://github.com/mingzi151/SpokenVocab