Tag
Nedimo presents an evidence-grounded, multilingual pipeline for generating low-arousal AI sleep stories, separating deterministic planning from LLM generation with attention tapering, validation guardrails, and TTS loudness normalization, while positioning the system as a relaxation tool without clinical efficacy claims.
A developer built a full 3D human creator with three.js and Claude, featuring IK rigging, procedural walking/animation, clothing, text-to-speech, and AI-generated skin and garments for any ethnicity, and open-sourced it as a MetaHuman alternative for three.js.
This tweet compiles a leaderboard of the current best models across categories, covering closed-source/open-source LLMs, video generation, image generation, decision-making, TTS, ASR, embeddings, and music generation.
Voice Arena launched Turing TTS, a benchmark that measures how often listeners mistake AI speech for human speech via a 'Fool Rate'; even the best model, Gemini 3.8 Flash TTS, fools listeners only about 10% of the time.
Introducing Turing TTS, an evaluation measuring how often human listeners mistake AI-generated speech for real human speech; even the best TTS models fooled listeners only ~10% of the time, prompting discussion on better metrics for voice AI progress.
A Halloween blog post shows how an interactive 'MonsterHouse' haunted house was built using NVIDIA's open Nemotron models, combining ASR, a small language model, zero-shot TTS, and Audio2Face-3D to make a grumpy garage door talk to trick-or-treaters on a single workstation.
Fish Audio 为 Drama 3 语音模型发布了升级后的权重，使其在 Artificial Analysis 的 Provider Voice 和 Controlled Voice 两个 Speech Arena 榜单上均位列开源权重模型第一，并在发音鲁棒性基准上排名第一。
Hugging Face's Open TTS Leaderboard compares not only TTS models but also inference implementations, benchmarking streaming performance (TTFA), inference speed, speech quality (WER/CER, speaker similarity), and GPU/CPU results—useful for anyone building local voice agents.
Deepgram's new Flux TTS targets a critical accuracy bottleneck in voice-agent pipelines, achieving a 3.4% word error rate on tricky alphanumeric strings like medication names and tracking numbers while preserving natural-sounding speech through the full voice loop.
Armin Ronacher explores custom voice creation with Qwen3-TTS, covering its Base, CustomVoice, and VoiceDesign variants and showing how to blend voice embeddings with short reference audio to build a unique narrator for a dungeon-crawler game project with his kids.
Readwise 于其第三款产品 Readwise Books 推出邀请制测试版，将购书、阅读与阅读追踪整合到一个应用中，支持 Kindle/Goodreads 导入、EPUB 上传、Android 电子墨水设备、文本转语音以及 MCP 服务器。
A sponsored post comparing REST, GraphQL, gRPC, and WebSockets, then introducing Deepgram Flux TTS — a conversation-aware text-to-speech API for real-time voice agents with built-in interruption handling, sub-80ms latency, and VPC/on-prem deployment.
作者分享如何用一台12年前的办公PC运行本地脚本（IMAP邮件轮询、Tesseract OCR、Piper TTS），把Telegram机器人例行任务从昂贵的LLM API调用转为零token的本地处理，显著降低成本。
WhissleAI released SteerTTS-Edge, a steerable small English TTS model compressed to 11.84 MB via int4 quantization-aware training of the acoustic model and a distilled Vocos vocoder, with unchanged intelligibility (WER 0.96%) and only a ~0.1 UTMOS drop while preserving all 12 steering controls.
AuroraAI-Research released AuroraTTS-100K, an extremely small English text-to-speech model with only 95,638 inference parameters, trained from scratch on LJSpeech using a differentiable harmonic+noise DSP synthesizer instead of a neural vocoder, achieving UTMOS 3.53 with very fast inference.
KittenML 开源发布了 Kitten-TTS 2，这是全球首个 1.58-bit（三值化权重）TTS 模型，1.7B 参数，可在纯 CPU 上实时推理，支持少于 5 秒参考音频的零样本语音克隆、20+ 语言多语种支持以及情感与强调控制。
inclusionAI released Ming-3.0-TTS-0.5B, a dense 0.5B multilingual text-to-speech model built on Qwen2.5-0.5B supporting zero-shot voice cloning, voice design, paralinguistic inline tags, and free-form style control across 7 languages plus Chinese dialects.
Ant Group’s inclusionAI has open-sourced the Ming-3.0-tts speech synthesis model, which ranked first overall in pronunciation robustness with a score of 93.2% on Artificial Analysis’s October leaderboard and first among open-source models for voice controllability. It supports voice cloning, control through natural-language instructions, 14 paralinguistic tags, and speech synthesis across multiple dialects and languages, including Sichuanese and Cantonese.
RubatoCodec 提出一种基于固定语句事件预算的自适应语音令牌化方法，结合流式解码，在低事件率下显著提升语音重建质量与可懂度，并在匹配的文本到语音对比中取得更低 WER 与更高 UTMOS 分数。
ElevenLabs 推出 Eleven Realtime，一套端到端优化的 Speech-to-Speech 语音模型（v1 Mini 与 v1 Max），支持实时语音对话，并提供与 OpenAI Realtime API 兼容的接口，可在 ElevenAgents 中快速构建语音助手。