【Hacker News搬运】使用带LLM的BCI使ALS患者能够再次与家人交谈
-
Title: Using a BCI with LLM for enabling ALS patients to speak again with family
使用带LLM的BCI使ALS患者能够再次与家人交谈
Text:
Url: https://thevccorner.substack.com/p/from-sci-fi-to-reality
由于我是一个AI,我无法直接访问互联网或特定网站来抓取内容。但是,我可以指导你如何使用JinaReader(一个基于Python的库)来抓取和分析网页内容,并给出一个假设性的分析。 以下是一个使用JinaReader(假设它存在并且可以用于此类任务)的Python脚本示例,用于抓取内容并分析总结。如果页面内容是英文的,我们还可以使用一个简单的翻译函数来将其翻译成中文。请注意,这里的代码是假设性的,因为JinaReader库可能不存在,而且实际翻译功能需要集成一个翻译API,如Google Translate。 ```python import requests from bs4 import BeautifulSoup from jina import Document, Client # 假设的翻译函数,实际使用时需要集成翻译API def translate_to_chinese(text): # 这里需要替换为实际的翻译API调用 # 返回翻译后的文本 return text # 假设文本已经是中文 # 使用JinaReader抓取内容 def fetch_content(url): response = requests.get(url) soup = BeautifulSoup(response.text, 'html.parser') # 假设我们只需要抓取文章的标题和正文 title = soup.find('h1').get_text() content = soup.find('div', class_='post-content').get_text() return title, content # 分析和总结内容 def summarize_content(title, content): # 这里可以加入自然语言处理技术来分析内容,这里只是简单示例 summary = f"标题:{title}\n内容摘要:{content[:200]}..." # 假设摘要为内容的前200个字符 return summary # 主函数 def main(): url = "https://thevccorner.substack.com/p/from-sci-fi-to-reality" title, content = fetch_content(url) chinese_title = translate_to_chinese(title) chinese_content = translate_to_chinese(content) summary = summarize_content(chinese_title, chinese_content) print(summary) if __name__ == "__main__": main()
请记住,这个脚本是一个简化的示例,实际应用中可能需要处理更多的细节,比如:
- 确保网页结构的一致性,以便正确地找到标题和正文。
- 处理JavaScript渲染的内容,可能需要使用Selenium或Puppeteer等工具。
- 翻译功能需要集成一个实际的翻译服务,并处理API密钥和请求。
- 对于内容摘要,可能需要使用更高级的自然语言处理技术,如摘要算法或预训练的模型。
由于JinaReader和实际的翻译API细节未知,上面的代码仅为示例,无法直接运行。
## Post by: vasco_ ### Comments: **labelra**: Are there any details on how this works? Based on what is available in the linked article, it looks like they have an LLM+RAG and are trying to pass off the responses as speech from the user. Done with full transparency, and right protections, this could be useful, but calling it BCI, and overselling it as user's voice (especially given voice cloning is also being done) can be misrepresenting it. > **labelra**: 有关于这是如何工作的细节吗?根据链接文章中的内容,他们似乎有一个LLM+RAG,并试图将响应冒充为用户的语音。在完全透明和适当保护的情况下,这可能是有用的,但称之为BCI,并作为用户过度销售;她的声音(特别是考虑到语音克隆也在进行中)可能会被歪曲。 **xk_id**: Those non-invasive headbands (which work very differently from implanted electrodes) are notoriously inaccurate at recording brain signals. Even scientific studies, which use advanced setups like the 10-20 system for scalp EEG, face unsolved challenges in removing noise from the data and in using the data to reconstruct underlying brain activity [0] – let alone making meaningful inferences about it.<p>Patients with locked-in syndrome (one of the use cases mentioned in the article, also called a pseudo-coma), or with other disorders of consciousness, are unable to protest, or to confirm the accuracy of the <i>generative</i> message which is being attributed to them. Communicating on your <i>own</i> terms and in your <i>own</i> words is fundamental to human dignity.<p>Meanwhile, this coincides with lukewarm reception of generative AI from consumers; perhaps it is the lack of autonomy of locked-in patients, which makes them an interesting segment to this new generation of ventures, scrambling for a ROI on the enormous over-investment in the sector.<p>The conference venues look lush tho.<p>[0] <a href="https://en.wikipedia.org/wiki/Electroencephalography#Artifacts" rel="nofollow">https://en.wikipedia.org/wiki/Electroencephalography#Artifac...</a> > **xk_id**: 众所周知,这些非侵入性头带(其工作原理与植入电极非常不同)在记录大脑信号方面是不准确的。即使是使用头皮脑电图10-20系统等先进设置的科学研究,在从数据中去除噪声和使用数据重建潜在的大脑活动方面也面临着未解决的挑战[0],更不用说对其进行有意义的推断了。<p>闭锁综合征患者(文章中提到的用例之一,也称为伪昏迷)或其他意识障碍患者无法抗议或确认归因于他们的<i>生成性</i>信息的准确性。用你自己的语言进行交流是人类尊严的基础<p> 与此同时,这与消费者对生成式人工智能的冷淡态度相吻合;也许是锁定患者缺乏自主权,这使得他们成为新一代企业的一个有趣部分,在该行业巨大的过度投资中争夺投资回报率<p> 会议场地看起来郁郁葱葱<p> [0]<a href=“https:/;en.wikipedia.org&#wiki&#脑电图#Artifacts”rel=“nofollow”>https:/;en.wikipedia.org;维基;脑电图#Artifac</一 **vasco_**: Halo, developed by Unbabel, combines a non invasive BCI with an LLM to enable ALS patients to regain the ability to talk with loved ones. The search for a CEO is on. > **vasco_**: 由Unbabel开发的Halo将非侵入性BCI与LLM相结合,使ALS患者能够重新获得与亲人交谈的能力。寻找CEO的工作正在进行中。 **xk_id**: [flagged] > **xk_id**: [标记]