Searcharxiv⌕ Search

arXiv subjects

Ryuichiro Higashinaka

Publications and source records attributed to Ryuichiro Higashinaka.

15 recordsLinked to original sources
arXiv2026

KABURI-TTS: Phoneme-Keyed Activity-conditioned Bi-channel Utterance Rendering for Interaction

Realizing full-duplex spoken dialogue requires large amounts of two-channel, one-speaker-per-channel conversational speech data. Although conversational text-to-speech (TTS) engines have been developed, they are not necessarily robust to two-party simultaneous phenomena such as backchannels, interruptions, and overlaps that occur while the interlocutor is speaking. In this work, aiming at conversational speech synthesis that reproduces human-like overlap, we propose KABURI-TTS. KABURI-TTS takes a per-speaker phoneme raster as input and renders the speech of the two speakers on separate channels, conditioned on the per-frame phonemes and the voice activity derived from them. Because the phoneme raster is supplied by a separate module, the proposed method enables controllable generation of one-speaker-per-channel, two-party spoken dialogue. A user evaluation shows that, compared with strong baselines, the proposed method attains higher naturalness at both the utterance and the interaction level. Furthermore, an analysis of voice activity confirms that the proposed method produces more overlap and more frequent turn-taking.

cs.SD↗
arXiv2026

Integrating Facial Generation into Full-Duplex Spoken Dialogue Systems

Full-duplex spoken dialogue models, such as Moshi, enable natural, low-latency voice conversations. However, they remain limited to the audio modality, lacking the facial expressions that are integral to human communication. We present Moshi-Face, the first full-duplex dialogue model that jointly processes the user's audio and facial input while simultaneously generating speech and facial motion. We first construct a vector-quantized variational autoencoder (VQ-VAE) as a face codec that encodes 3D head meshes extracted from facial videos into compact discrete tokens, referred to as face tokens, and conversely reconstructs 3D meshes from these tokens. We then extend Moshi with a Face Transformer module that generates face tokens non-autoregressively, enabling Moshi-Face to produce synchronized audio and face tokens in real time. Experiments show that Moshi-Face achieves audiovisual alignment at low latency while preserving the dialogue quality of the original audio-only model.

cs.HC↗
arXiv2026

Llama-Mimi: Exploring the Limits of Flattened Speech Language Modeling

Speech Language Models (SpeechLMs) model tokenized speech to capture both semantic and acoustic information. When neural audio codecs based on Residual Vector Quantization (RVQ) are used as audio tokenizers, they produce multiple discrete tokens per time step, yielding inherently multi-level representations. To process these multi-level tokens together, prior work typically adopts hierarchical architectures to capture this structure. In contrast, recent progress in NLP has progressively reduced architectural inductive biases, moving toward simpler and more scalable single-Transformer architectures. In this work, we propose Llama-Mimi, which flattens multi-level RVQ tokens produced by the Mimi neural audio codec into a single sequence and models them autoregressively with a Transformer decoder. We show that Llama-Mimi outperforms a CSM-based hierarchical model on most tasks and achieves the best performance on acoustic consistency. Our models, code, and speech samples are publicly available.

cs.CL↗
arXiv2025

Towards a Japanese Full-duplex Spoken Dialogue System

Full-duplex spoken dialogue systems, which can model simultaneous bidirectional features of human conversations such as speech overlaps and backchannels, have attracted significant attention recently. However, the study of full-duplex spoken dialogue systems for the Japanese language has been limited, and the research on their development in Japanese remains scarce. In this paper, we present the first publicly available full-duplex spoken dialogue model in Japanese, which is built upon Moshi, a full-duplex dialogue model in English. Our model is trained through a two-stage process: pre-training on a large-scale spoken dialogue data in Japanese, followed by fine-tuning on high-quality stereo spoken dialogue data. We further enhance the model's performance by incorporating synthetic dialogue data generated by a multi-stream text-to-speech system. Evaluation experiments demonstrate that the trained model outperforms Japanese baseline models in both naturalness and meaningfulness.

cs.CL↗
arXiv2025

Universal Post-Processing Networks for Joint Optimization of Modules in Task-Oriented Dialogue Systems

Post-processing networks (PPNs) are components that modify the outputs of arbitrary modules in task-oriented dialogue systems and are optimized using reinforcement learning (RL) to improve the overall task completion capability of the system. However, previous PPN-based approaches have been limited to handling only a subset of modules within a system, which poses a significant limitation in improving the system performance. In this study, we propose a joint optimization method for post-processing the outputs of all modules using universal post-processing networks (UniPPNs), which are language-model-based networks that can modify the outputs of arbitrary modules in a system as a sequence-transformation task. Moreover, our RL algorithm, which employs a module-level Markov decision process, enables fine-grained value and advantage estimation for each module, thereby stabilizing joint learning for post-processing the outputs of all modules. Through both simulation-based and human evaluation experiments using the MultiWOZ dataset, we demonstrated that UniPPN outperforms conventional PPNs in the task completion capability of task-oriented dialogue systems.

cs.CL↗
arXiv2024

JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset

Dialogue datasets are crucial for deep learning-based task-oriented dialogue system research. While numerous English language multi-domain task-oriented dialogue datasets have been developed and contributed to significant advancements in task-oriented dialogue systems, such a dataset does not exist in Japanese, and research in this area is limited compared to that in English. In this study, towards the advancement of research and development of task-oriented dialogue systems in Japanese, we constructed JMultiWOZ, the first Japanese language large-scale multi-domain task-oriented dialogue dataset. Using JMultiWOZ, we evaluated the dialogue state tracking and response generation capabilities of the state-of-the-art methods on the existing major English benchmark dataset MultiWOZ2.2 and the latest large language model (LLM)-based methods. Our evaluation results demonstrated that JMultiWOZ provides a benchmark that is on par with MultiWOZ2.2. In addition, through evaluation experiments of interactive dialogues with the models and human participants, we identified limitations in the task completion capabilities of LLMs in Japanese.

cs.CL↗
arXiv2024

Proceedings of the Dialogue Robot Competition 2023

The Dialogic Robot Competition 2023 (DRC2023) is a competition for humanoid robots (android robots that closely resemble humans) to compete in interactive capabilities. This is the third year of the competition. The top four teams from the preliminary competition held in November 2023 will compete in the final competition on Saturday, December 23. The task for the interactive robots is to recommend a tourism plan for a specific region. The robots can employ multimodal behaviors, such as language and gestures, to engage the user in the sightseeing plan they recommend. In the preliminary round, the interactive robots were stationed in a travel agency office, where visitors conversed with them and rated their performance via a questionnaire. In the final round, dialogue researchers and tourism industry professionals interacted with the robots and evaluated their performance. This event allows visitors to gain insights into the types of dialogue services that future dialogue robots should offer. The proceedings include papers on dialogue systems developed by the 12 teams participating in DRC2023, as well as an overview of the papers provided by all the teams.

cs.RO↗
arXiv2024

Overview of Dialogue Robot Competition 2023

We have held dialogue robot competitions in 2020 and 2022 to compare the performances of interactive robots using an android that closely resembles a human. In 2023, the third competition DRC2023 was held. The task of DRC2023 was designed to be more challenging than the previous travel agent dialogue tasks. Since anyone can now develop a dialogue system using LLMs, the participating teams are required to develop a system that effectively uses information about the situation on the spot (real-time information), which is not handled by ChatGPT and other systems. DRC2023 has two rounds, a preliminary round and the final round as well as the previous competitions. The preliminary round has held on Oct.27 -- Nov.20, 2023 at real travel agency stores. The final round will be held on December 23, 2023. This paper provides an overview of the task settings and evaluation method of DRC2023 and the preliminary round results.

cs.RO↗
arXiv2023

Team Flow at DRC2023: Building Common Ground and Text-based Turn-taking in a Travel Agent Spoken Dialogue System

At the Dialogue Robot Competition 2023 (DRC2023), which was held to improve the capability of dialogue robots, our team developed a system that could build common ground and take more natural turns based on user utterance texts. Our system generated queries for sightseeing spot searches using the common ground and engaged in dialogue while waiting for user comprehension.

cs.CL↗
arXiv2023

Cognitive Architecture Toward Common Ground Sharing Among Humans and Generative AIs: Trial on Model-Model Interactions in Tangram Naming Task

For generative AIs to be trustworthy, establishing transparent common grounding with humans is essential. As a preparation toward human-model common grounding, this study examines the process of model-model common grounding. In this context, common ground is defined as a cognitive framework shared among agents in communication, enabling the connection of symbols exchanged between agents to the meanings inherent in each agent. This connection is facilitated by a shared cognitive framework among the agents involved. In this research, we focus on the tangram naming task (TNT) as a testbed to examine the common-ground-building process. Unlike previous models designed for this task, our approach employs generative AIs to visualize the internal processes of the model. In this task, the sender constructs a metaphorical image of an abstract figure within the model and generates a detailed description based on this image. The receiver interprets the generated description from the partner by constructing another image and reconstructing the original abstract figure. Preliminary results from the study show an improvement in task performance beyond the chance level, indicating the effect of the common cognitive framework implemented in the models. Additionally, we observed that incremental backpropagations leveraging successful communication cases for a component of the model led to a statistically significant increase in performance. These results provide valuable insights into the mechanisms of common grounding made by generative AIs, improving human communication with the evolving intelligent machines in our future society.

cs.AI↗
arXiv2022

Overview of Dialogue Robot Competition 2022

Although many competitions have been held on dialogue systems in the past, no competition has been organized specifically for dialogue with humanoid robots. As the first such attempt in the world, we held a dialogue robot competition in 2020 to compare the performances of interactive robots using an android that closely resembles a human. Dialogue Robot Competition 2022 (DRC2022) was the second competition, held in August 2022. The task and regulations followed those of the first competition, while the evaluation method was improved and the event was internationalized. The competition has two rounds, a preliminary round and the final round. In the preliminary round, twelve participating teams competed in performance of a dialogue robot in the manner of a field experiment, and then three of those teams were selected as finalists. The final round will be held on October 25, 2022, in the Robot Competition session of IROS2022. This paper provides an overview of the task settings and evaluation method of DRC2022 and the results of the preliminary round.

cs.RO↗
arXiv2022

Team Flow at DRC2022: Pipeline System for Travel Destination Recommendation Task in Spoken Dialogue

To improve the interactive capabilities of a dialogue system, e.g., to adapt to different customers, the Dialogue Robot Competition (DRC2022) was held. As one of the teams, we built a dialogue system with a pipeline structure containing four modules. The natural language understanding (NLU) and natural language generation (NLG) modules were GPT-2 based models, and the dialogue state tracking (DST) and policy modules were designed on the basis of hand-crafted rules. After the preliminary round of the competition, we found that the low variation in training examples for the NLU and failed recommendation due to the policy used were probably the main reasons for the limited performance of the system.

cs.CL↗
arXiv2022

Adaptive Natural Language Generation for Task-oriented Dialogue via Reinforcement Learning

When a natural language generation (NLG) component is implemented in a real-world task-oriented dialogue system, it is necessary to generate not only natural utterances as learned on training data but also utterances adapted to the dialogue environment (e.g., noise from environmental sounds) and the user (e.g., users with low levels of understanding ability). Inspired by recent advances in reinforcement learning (RL) for language generation tasks, we propose ANTOR, a method for Adaptive Natural language generation for Task-Oriented dialogue via Reinforcement learning. In ANTOR, a natural language understanding (NLU) module, which corresponds to the user's understanding of system utterances, is incorporated into the objective function of RL. If the NLG's intentions are correctly conveyed to the NLU, which understands a system's utterances, the NLG is given a positive reward. We conducted experiments on the MultiWOZ dataset, and we confirmed that ANTOR could generate adaptive utterances against speech recognition errors and the different vocabulary levels of users.

cs.CL↗
arXiv2022

Post-processing Networks: Method for Optimizing Pipeline Task-oriented Dialogue Systems using Reinforcement Learning

Many studies have proposed methods for optimizing the dialogue performance of an entire pipeline task-oriented dialogue system by jointly training modules in the system using reinforcement learning. However, these methods are limited in that they can only be applied to modules implemented using trainable neural-based methods. To solve this problem, we propose a method for optimizing a pipeline system composed of modules implemented with arbitrary methods for dialogue performance. With our method, neural-based components called post-processing networks (PPNs) are installed inside such a system to post-process the output of each module. All PPNs are updated to improve the overall dialogue performance of the system by using reinforcement learning, not necessitating each module to be differentiable. Through dialogue simulation and human evaluation on the MultiWOZ dataset, we show that our method can improve the dialogue performance of pipeline systems consisting of various modules.

cs.CL↗