1
阿里推出国内首个 AI 语音平台 CosyVoice Studio，限时免费体验
一站式 AI 语音生产力平台，自研语音模型加持，还能自动过滤口头语，限时免费体验。
IT之家 8 月 7 日消息，阿里巴巴今日宣布推出国内首个一站式 AI 语音生产力平台 CosyVoice Studio 。 官方表示，该平台基于自研语音模型 Qwen-Audio 打造，包含语音记录 CosyFlow、语音智能体 CosyAgent 和音频内容创作工具 CosyCreative 等…
2
Controllable Accent Normalization via Discrete Diffusion
口音归一化新研究，离散扩散模型实现可控语音转换，论文思路前沿，值得语音方向开发者深挖。
arXiv:2603.14275v2 Announce Type: replace-cross Abstract: Existing accent normalization methods do not typically offer control over accent strength, y…
3
Fair Cognitive Impairment Detection Through Unlearning
用「去学习」技术消除模型偏见，实现公平的认知障碍检测，AI医疗伦理前沿探索。
arXiv:2606.18571v1 Announce Type: new Abstract: Mild Cognitive Impairment (MCI) is a medical condition characterized by a noticeable decline in memory…
4
Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data
利用Audio-LLMs智能过滤低质量数据，大幅提升语音到语音翻译模型的训练效率与效果，INTERSPEECH 2026录用论文。
arXiv:2606.13507v1 Announce Type: new Abstract: Large-scale mined corpora provide abundant training data for end-to-end speech-to-speech translation (…
5
Whisper-GPT -- Continuous Discrete Hybrid Representation Language Models For Speech And Music
一种新颖的连续离散混合表示语言模型，同时处理语音与音乐，统一建模音频生成和理解。
arXiv:2412.11449v2 Announce Type: replace-cross Abstract: We propose WHISPER-GPT: A generative large language model (LLM) for speech and music that al…
6
Show HN: Entropic — information-driven variable-rate media playback
基于信息熵的智能变速播放工具，自动跳过低信息片段，提升听觉效率。
Article URL: https://github.com/patrickxia/entropic Comments URL: https://news.ycombinator.com/item?id=48346255 Points: 4 # Comments: 1