13 papers
MSVAE: Rethinking Musical Semantics in Continuous Latent Spaces
Guobin Ma, Yuepeng Jiang, Dake Guo +15
Continuous latent representations underlie recent diffusion-based lyrics-to-song and text-to-music systems, yet aligning lyric conditions with these latents remains difficult: exis…
StreamTN: A Low-Latency Streaming Chinese Text Normalization Model for Streaming TTS in Dialogue Systems
Wenhao Li, Jinrui Liang, Haoyu Zhang +11
Text-to-Speech (TTS) is an essential module that provides spoken responses in a spoken dialogue system (SDS) centered on a large language model (LLM). To ensure accurate TTS synthe…
Rethinking Music Tokenization: A Semantic Codec toward High-Fidelity LLM Music Generation
Huakang Chen, Guobin Ma, Yuepeng Jiang +10
Discrete audio tokenization has become the critical interface between raw waveforms and autoregressive modeling in recent music generation. As a result, music tokenizers must simul…
SphereVAE: Hyperspherical Latent Autoencoders for Robust Autoregressive Speech Representation Modeling
Haoyu Zhang, Jingbin Hu, Hanke Xie +8
With the rapid development of speech generation technology, discrete codec representations have been widely used because they provide a stable prediction paradigm. In expressive sp…
SemBridge: Semantic Token Anchoring for Continuous-Latent Autoregressive Speech Generation
Hanke Xie, Haopeng Lin, Jiale Qian +13
Continuous-latent autoregressive speech generation has emerged as a promising alternative to discrete-token modeling by avoiding quantization loss and preserving richer acoustic in…
MeanVC 2: Robust Low-Latency Streaming Zero-Shot Voice Conversion
Guobin Ma, Yuxuan Xia, Yuepeng Jiang +6
Streaming zero-shot voice conversion (VC) has become increasingly popular due to its potential for real-time applications. The recently proposed MeanVC achieves lightweight streami…
G-MaP-SE: Guided Speech Enhancement via GMM-Based Prior Matching
Yike Zhu, Ziqian Wang, Zikai Liu +5
Using speaker embeddings as conditioning can strengthen speech enhancement, but most methods either require clean enrollment audio or rely on embeddings extracted from noisy speech…
Joint Learning Global-Local Speaker Classification to Enhance End-to-End Speaker Diarization and Recognition
Yuhang Dai, Haopeng Lin, Jiale Qian +9
Large Audio-Language Models (LALMs) have demonstrated remarkable performance in end-to-end speaker diarization and recognition. However, their speaker discriminability remains limi…
OmniCodec: Low Frame Rate Universal Audio Codec with Semantic-Acoustic Disentanglement
Jingbin Hu, Haoyu Zhang, Dake Guo +10
Large Language Models (LLMs) have advanced audio generation through discrete representation learning. However, most existing neural codecs focus on speech and emphasize reconstruct…
SoulX-Podcast: Towards Realistic Long-form Podcasts with Dialectal and Paralinguistic Diversity
Hanke Xie, Haopeng Lin, Wenxiao Cao +17
Recent advances in text-to-speech (TTS) synthesis have significantly improved speech expressiveness and naturalness. However, most existing systems are tailored for single-speaker…
ContextASR-Bench: A Massive Contextual Speech Recognition Benchmark
He Wang, Linhan Ma, Dake Guo +4
Automatic Speech Recognition (ASR) has been extensively investigated, yet prior benchmarks have largely focused on assessing the acoustic robustness of ASR models, leaving evaluati…
StreamFlow: Streaming Flow Matching with Block-wise Guided Attention Mask for Speech Token Decoding
Dake Guo, Jixun Yao, Linhan Ma +2
Recent advancements in discrete token-based speech generation have highlighted the importance of token-to-waveform generation for audio quality, particularly in real-time interacti…
FlexSpeech: Towards Stable, Controllable and Expressive Text-to-Speech
Linhan Ma, Dake Guo, He Wang +2
Current speech generation research can be categorized into two primary classes: non-autoregressive and autoregressive. The fundamental distinction between these approaches lies in…