Tag
The developer of Subtitle Foundry explains why keeping ASR, translation, and timing errors separately inspectable matters in a subtitle pipeline, even when using SOTA speech and language tools. The post outlines three distinct checks, evaluation practices across providers (including ElevenLabs Scribe v2 word timestamps), and links the AGPL-3.0 project.
An arXiv case study examining how professional translators can achieve digital sovereignty by migrating their workflows from Windows to Linux.
WIRED's roundup of the best handheld translator devices for work and travel in 2026, covering standalone translators, earbud-based translators, and how they compare to free phone apps like Google Translate and DeepL.
StoreLocale 是一款在 Product Hunt 上发布的工具，可自动检测并翻译 App Store 与 Google Play 截图中的 UI 文本，支持 32 种语言并保持原有布局，编辑器免费、可通过自带 API Key 或一次性积分使用 AI 图像生成。
IBM 发布 Granite Speech 5.0 系列更新，包括新的 2B 语音语言模型以及改进的 470m-turboctc 和 4.1-2b-plus 版本，在转录、翻译、抗噪和说话人归属方面更准确快速，采用 Apache 2.0 许可。
ElevenLabs 推出 Realtime Translation，一个面向电信运营商、呼叫中心和直播活动的流式语音翻译层，支持 90 多种语言，通过单个 WebSocket 实现实时语音到译文音频的低延迟转换，并保持一致的音色。
A user complains that streaming services like Amazon Prime are replacing accurate, human-made subtitles on older shows with error-ridden AI-generated subtitles, questioning why existing high-quality captions aren't reused.
NVIDIA 与 AI4Bharat/IIT Madras 联合发布 Nemotron-pretraining-Indic-v1，涵盖 22 种印度语言的约 2.8 万亿 token 预训练语料，并开放可复用的 GPU 加速数据生成流水线。
A guide shared by an Anthropic Claude team member (English translation), proposing that we raise our ambition when using AI along four dimensions — speed, altitude, depth, and range — with practical methods and prompt techniques.
PixLate 是一款在 Product Hunt 上发布的文档翻译工具，可在保留原有排版与插图的情况下翻译 PDF、扫描件和图文页面，并提供并排校对、后台处理与本地化隐私处理，采用按次计费模式。
ArXivHub 是一个开源工具，输入 arXiv ID 即可一站式跳转到 116 个阅读、AI 解读、翻译、代码检索与讨论入口，同时提供网页版和 Tampermonkey 浏览器脚本。
推荐 FluentRead，一个开源浏览器翻译插件，具备双语网页翻译、AI 阅读卡片、图片与文档翻译及双语视频字幕等功能，并集成了 DeepSeek Harness 会话核心用于上下文解释与追问。
Typeling 是一款在 iPhone 键盘内直接翻译的应用，支持六种语言和多种语气（朋友、伴侣、商务），可预览译文后一键替换，也能粘贴外语消息进行翻译。
NihonSub is an open-source real-time Japanese anime subtitle and translation engine combining Whisper-Large-v3 on Groq with DeepSeek/LLaMA-3 for contextual bilingual subtitles, featuring VAD-based audio chunking, honorific-preserving LLM translation, and a synchronized dual-subtitle cinema player.
Jarq is a macOS utility launched on Product Hunt that lets users translate, shorten, fix, or rewrite text in-place anywhere on their Mac via hotkeys, plus cross-language voice dictation in 13 languages. It uses cloud AI models without storing text on servers, with a free tier of 5 actions per day and Pro at $4.99/mo or $39/yr.
mlsubgen is an open-source CLI tool that generates video subtitles in 45 languages entirely on a local machine, using CUDA-based ASR (Qwen3-ASR, Whisper large-v3) and Ollama-hosted LLM translation with hardware profiles down to 8 GB GPUs.
Voicera launches on Product Hunt as a free AI-powered tool that converts natural speech into polished text and lets users rewrite, fix, and translate their writing in place.
Bilibili's Index LLM team releases Index-Translate and companion models (Index-NativeLong, Index-Homura, Index-Echo) supporting 150 text languages, document translation, multilingual subtitles, and voice-preserving dubbing, released under Apache-2.0 with open weights in 2B, 9B, and 35B-A3B sizes.
Vitra.ai launched on Product Hunt as an agentic content platform that consolidates 12 tools into one workspace, letting AI agents create, translate, and localize video, images, docs, and audio across 100+ languages with lip-sync dubbing and brand consistency.
BiliBili Index team announced Index-Translate, a family of multilingual translation models (2B/9B/35A3B) built on Qwen3.5 supporting 150 languages, alongside speech-to-text, speech-to-speech, syllable-constrained, and long-document translation variants released under Apache 2.0.