LLM时代语音合成与转换加剧深度伪造风险，这项新基准测试揭露了现有欺骗检测器的泛化短板。
arXiv:2607.11706v1 Announce Type: cross Abstract: Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems produce synthetic speech th…
为妈妈打造的Rofi Linux助手，解决远程技术支持烦恼，兼顾实用与温情。
Honestly, this wasn't supposed to become a project. There are already a few AI desktop assistants built around Rofi. They work, but they usually cover…
开源AI语音工作室，集成多款TTS模型，支持zero-shot声音克隆和MCP智能体语音交互。
Article URL: https://github.com/jamiepine/voicebox Comments URL: https://news.ycombinator.com/item?id=48863385 Points: 1 # Comments: 0
卢森堡语口语问答新突破，用TTS增强数据实现低资源语言语音交互。
arXiv:2607.02763v1 Announce Type: new Abstract: Spoken Question Answering (SQA) remains largely focused on high-resource languages and carefully recor…
开源电话呼叫基础设施，帮AI代理打通语音通话能力，集成STT、LLM、TTS全链路。
Article URL: https://github.com/AgentLineHQ/AgentLine Comments URL: https://news.ycombinator.com/item?id=48793076 Points: 3 # Comments: 3
这项研究通过数据扩展与定向合成，攻克了日语语音生成中汉字多音词的难题。
arXiv:2606.25369v1 Announce Type: cross Abstract: While large language model (LLM)-based text-to-speech (TTS) systems have achieved high-quality speec…
开源语音AI平台Omni-VRAM集成了28个Python模块，支持语音识别、实时流、情绪识别等，<200ms延迟，免费可用！
What I Built Omni-VRAM is an open-source voice AI platform with 28 modules. GitHub: https://github.com/Liangchenxu/Omni-VRAM Features Speech Recogniti…
当AI语音逼真到难以分辨，未来的有声读物和交互方式将迎来巨变，HN社区热议技术成熟后的社会影响。
I fell into the rabbit hole of TTS models lately. Tried all major paid tools (ElevenLabs/InWorld/etc.), and all the newest open-source models. I start…
基于Plutchik情感模型的价值规划方法，实现高表现力流式情感语音合成，被ACL 2026录用。
arXiv:2606.09837v1 Announce Type: cross Abstract: Emotional interaction is increasingly crucial for conversational AI, yet current systems lack a self…
提出端到端训练方法，将离散token LLM与TTS系统结合，提升语音合成效果。
arXiv:2606.09234v1 Announce Type: cross Abstract: Recent state-of-the-art (SOTA) text-to-speech (TTS) systems typically adopt a cascaded pipeline cons…
开源新SOTA语音合成模型，4B参数，Boson AI发布Higgs Audio v3-TTS，音质飞跃。
Article URL: https://huggingface.co/bosonai/higgs-audio-v3-tts-4b Comments URL: https://news.ycombinator.com/item?id=48409001 Points: 3 # Comments: 0
微软推出支持15种语言的带情绪表达语音克隆TTS新工具
Expressive TTS with voice cloning in 15 languages Discussion | Link
多语言语音合成工具，支持极低39ms延迟，在多语种TTS领域表现突出，适合需要高质量语音生成的场景。
I originally started lurking around on HN to better understand how startups work. I just saw the KugelAudio (YC P26) launch announcement and I truly d…
开源语音代理Knotch，基于圆盘辐条架构，实现动态对话管理与多参与者协调
Article URL: https://github.com/akshatvasisht/knotch Comments URL: https://news.ycombinator.com/item?id=48362727 Points: 3 # Comments: 1
访问Chatterbox-Flash论文，了解优先校准块扩散在零样本流式TTS中的前沿应用
arXiv:2605.30748v1 Announce Type: cross Abstract: We present Chatterbox-Flash, a zero-shot text-to-speech model obtained by fine-tuning a pretrained a…
副语言标签不出声？试试这个TTS，情感表达精准到位，告别静音读字
If you've worked with expressive TTS in the last year you've probably seen the pattern: She paused. [sigh] "Fine, you can come in." Inline paralinguis…
阶跃星辰一次性开源多款AI模型，Step 3.7 Flash生成速度高达每秒400 Tokens，刷新行业效率标杆。
IT之家 5 月 29 日消息，阶跃星辰今日正式发布并开源 Step 3.7 Flash，是面向 Agent 生产化阶段推出的新一代 Flash 模型，围绕 Agent、Coding、Search 与多模态工作流进行系统优化。 Step 3.7 Flash 采用稀疏 MoE 架构， 总参数 196B…