DailyTalk

The 'DailyTalk' dataset is a benchmark that contains multimodal dialogue history data used to evaluate conversational speech synthesis systems by analyzing the interaction between text and speech modalities.

Papers using DailyTalk (3)

DailyTalk: Spoken Dialogue Dataset for Conversational Text-to-Speech2022 · 3 citesParalinguistics-Aware Speech-Empowered Large Language Models for Natural Conversation2024 · 1 citesIntra- and Inter-modal Context Interaction Modeling for Conversational Speech Synthesis2024