论文解读
把拉格从玄学拉回有限状态：当重建变成约束 MAP 的精确动态规划
音乐理解 | 6.8/10
自回归模型
论文解读
从一次性生成到先诊断再重做：LoopTTS 如何让语音大模型当韵律医生
语音合成 | 7.0/10
论文解读
别等整句说完：把数字人的手势变成一条受因果约束的数据流
音视频交互 | 6.9/10
论文解读
X2Streaming-TTS: Causal Token-Level Text-to-Speech from Streaming Text with Speech-State Inheritance
语音合成 | 7.9/10
论文解读
Architecture and Affordances of PLAUD: Performative Latents and Unsupervised DDSP
音乐生成 | 5.0/10
论文解读
Omni-LiveAvatar: Minute-Level Real-Time Streaming Joint Audio-Visual Avatar Generation
音视频生成 | 7.7/10
论文解读
Alignment Drift in Single-Model Speculative Decoding for ASR: Mechanism, Correction, and Cost
语音识别 | 7.3/10
论文解读
VoxAudio: Vocalized Audio Synthesis via Multi-Reward Autoregressive Flow Matching
音频生成 | 8.2/10
论文解读
Confucius4-TTS: Transcript-Free Cross-Lingual Zero-Shot TTS with a Learnable Speaker Encoder
语音合成 | 7.7/10
论文解读
Phoenix TTS: High-Fidelity Synthesis and Voice Conversion via Flow-Matching-Driven Speech Tokenization
语音合成 | 6.5/10
论文解读
A Unifying Perspective on Audio Generative Modeling: Latent Representations and Modeling Strategies
音频生成 | 6.5/10
论文解读
CtrlSpeech: Coarse-to-Fine Control for Expressive Speech Synthesis
语音合成 | 6.4/10
论文解读
CuteTTS: Efficient and High-Quality Speech Synthesis via Autoregressive Modeling of Continuous Latents
语音合成 | 7.0/10
论文解读
MusicLayout: Explicit Structural Planning for Controllable Text-to-Music Generation
音乐生成 | 6.3/10
论文解读
MI-MIDI: Mechanistic Interpretability of Text-to-MIDI Generation Models via Probing, Lenses and Steering
音乐生成 | 7.1/10
论文解读
SemBridge: Semantic Token Anchoring for Continuous-Latent Autoregressive Speech Generation
语音合成 | 7.0/10
论文解读
Diff-Symbo: Text-Controlled Long-Duration Symbolic Music Generation Using Autoregressive Latent Diffusion Model
音乐生成 | 6.5/10
论文解读
Calliphony: A Calligraphy-Driven Interface for Real-Time Generative Music Performance
音乐生成 | 5.0/10
论文解读
CLASVS: Continuous-Latent Autoregression for Melody-Preserving Lyric Editing in Singing Voice Synthesis
自回归模型 | 7.1/10
论文解读
dots.tts.edit: Precisely Controlled Speech Editing with a Continuous Autoregressive Model
语音编辑 | 7.4/10