Publications (8)
The Conversational Short-phrase Speaker Diarization (CSSD) Task: Dataset, Evaluation Metric and Baselines
Gaofeng Cheng, Yifan Chen, Runyan Yang +9
The conversation scenario is one of the most important and most challenging scenarios for speech processing technologies because people in conversation respond to each other in a c…
PolySpeech: Exploring Unified Multitask Speech Models for Competitiveness with Single-task Models
Runyan Yang, Huibao Yang, Xiqing Zhang +6
Recently, there have been attempts to integrate various speech processing tasks into a unified model. However, few previous works directly demonstrated that joint optimization of d…
B-GRPO: Unsupervised Speech Emotion Recognition based on Batched-Group Relative Policy Optimization
Yingying Gao, Shilei Zhang, Runyan Yang +2
Unsupervised speech emotion recognition (SER) focuses on addressing the problem of data sparsity and annotation bias of emotional speech. Reinforcement learning (RL) is a promising…
Teaching Audio Models to Reason: A Unified Framework for Source- and Layer-wise Distillation
Runyan Yang, Yuke Si, Yingying Gao +3
While large audio language models excel at tasks like ASR and emotion recognition, they still struggle with complex reasoning due to the modality gap between audio and text as well…
Open Source MagicData-RAMC: A Rich Annotated Mandarin Conversational(RAMC) Speech Dataset
Zehui Yang, Yifan Chen, Lei Luo +9
This paper introduces a high-quality rich annotated Mandarin conversational (RAMC) speech dataset called MagicData-RAMC. The MagicData-RAMC corpus contains 180 hours of conversatio…
HarmoniFuse: A Component-Selective and Prompt-Adaptive Framework for Multi-Task Speech Language Modeling
Yuke Si, Runyan Yang, Yingying Gao +3
Recent advances in large language models have facilitated the development of unified speech language models (SLMs) capable of supporting multiple speech tasks within a shared archi…
DeCodec: Rethinking Audio Codecs as Universal Disentangled Representation Learners
Xiaoxue Luo, Jinwei Huang, Runyan Yang +4
Universal audio codecs learn entangled representations across audio types, whereas some specific codecs offer decoupled representations but are limited to speech. Real-world audio,…
Summary on the ISCSLP 2022 Chinese-English Code-Switching ASR Challenge
Shuhao Deng, Chengfei Li, Jinfeng Bai +6
Code-switching automatic speech recognition becomes one of the most challenging and the most valuable scenarios of automatic speech recognition, due to the code-switching phenomeno…