TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Swedish Royal Library releases speech-to-text-model

1 点作者 napoleongl3 个月前

1 comment

napoleongl3 个月前
Being a friend of one of the developers I’ve heard about this while it was in the works, but now it is officially released. 50k hours of parliament debates and other sources are at the foundation of it. Article in Swedish only but I guess the crowd here can solve that. Hopefully it becomes the standard model in places like MS Teams as their current model can’t handle dialects particularly well.