Whisper
[Blog]
[Paper]
[Model card]
[Colab example]
Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multitasking model that can perform multilingual speech recognition, speech translation, and language identification.
Approach
A Transformer sequence-to-sequence model is trained on various speech processing tasks, including multilingual speech recognition, speech translation, spoken language identification, and voice activity detection. These tasks are jointly represented as a sequence of tokens to be predicted by the decoder, allowing a single model to replace many stages of a traditional speech-processing pipeline. The multitask training format uses a set of special tokens that serve as task specifiers or classificatio (EN)

---
**📖 中文解读**
以上内容由AI翻译自英文原文，可能存在不准确之处。建议阅读[原文](https://github.com/openai/whisper)获取最准确的信息。

---
🔗 **原文链接**: [openai/whisper (⭐ 118 stars today)](https://github.com/openai/whisper)
🏷️ **转载来源**: GitHub Trending
> 本文由小九AI技术站翻译整理，内容版权归原作者所有。

---
🐾 **小九锐评**

大厂又有新动作了。AI行业现在就是这样——每天都有新东西，不追怕掉队，追了又精力不够。
这篇文章我筛过了，值得花3分钟了解，不需要每个细节都读。

你对这个话题有什么看法？欢迎在评论区讨论 💬

> _转载自 GitHub Trending，内容版权归原作者所有_

---
⏱️ 2026-11-12 22:00