There are hundreds of applications for generative AI after ChatGPT made such tools popular. Recently, Aiello, which is committed to building an AI voice open platform, combined its own technology with the currently popular OpenAI GPT large-scale language model to create Vocol.ai, a service that mainly uses AI to generate verbatim transcripts of voice conversations and perform topic summaries. Continue reading the free and easy-to-use GPT AI verbatim + meeting key analysis tool Vocol.ai tutorial, available in Chinese, English and Japanese! Report text.
The free and easy-to-use GPT AI verbatim draft + meeting key analysis tool Vocol.ai provides instruction in both Chinese, English and Japanese!
The application of generative AI can be said to be flourishing after ChatGPT brought the popularity of such tools. Recently, Aiello, which is committed to building an AI voice open platform, combined its own technology with the currently popular GPT large-scale language model to create Vocol.ai, a service that mainly uses AI to generate verbatim transcripts of voice conversations and perform topic summaries.
👉 Click here to try Vocol AI verbatim service (portal)
In fact, it is very similar to the verbatim draft and meeting arrangement service functions we introduced earlier. However, it integrates the flexibility of uploading voice files and even video files, and in addition to being able to quickly play voice clips by clicking on verbatim text, it can also be downloaded as files including srt / docx / csv, etc. In addition, it also provides integration functions for collaboration and direct connection with Microsoft Teams and Google Meet to transcribe meeting content – recording directly through the web interface is also possible. It is currently a fairly comprehensive AI voice-to-transcript service.
The point is, Vocol.ai also generously provides approximately 200 minutes of V-points for free trial – which is really generous, because the author of this tutorial uploaded three videos totaling nearly an hour, but has not consumed half of it. I highly recommend everyone to give it a try. Overall, the number of free minutes is quite sufficient. You can try out their features to your heart’s content.
The Vocol.ai interface is quite simple. After you simply sign in with a Microsoft or Google account to apply for an account, you will be able to master their interface through simple tutorials.
You can upload meeting minutes or any files you want to convert into verbatim transcripts through their service to this service through the “Upload File” in the upper right corner; to the left of the upload button is a “Start Recording” button for direct recording. The language part of the interface and analysis generation supports Traditional Chinese, English and Japanese. So for users in Taiwan, this is a very friendly service – after all, it is made by a Taiwanese development team, so we can really support it.
I believe that what everyone should be most curious about is the effect of this service. Although we did not have the meeting materials at hand, we tried to transcribe the interviews with Elon Musk and the videos on Ada’s YouTube. To be honest, although I feel that there are still some areas that need to be adjusted in the Vocol.ai interface (the team mentioned that many new features are being prepared), the effect is actually quite amazing.
Basically, after Vocol.ai uploads the verbatim transcript of the video file, whether it is a 10-minute Chinese content for a simple trial or an English interview of more than 20 minutes, their AI can complete the transcription in less than 5 minutes (actually it should be about 2 minutes…).
The point is that everything is basically done through the cloud, so after uploading the file, you can basically close the web page and wait for their email notification, and then go back to the interface to view the results of the verbatim draft. No need to wait in front of the computer for processing (but you have to wait until the transmission is completed!).
The best thing about Vocol.ai is that it not only organizes voice data into an online database list. After clicking on it, you can find that it does more than just help you complete the verbatim transcription. The webpage will also provide the topic and summary content of the entire conversation (if the name of the “speaker” is edited, the analyzed content will also be filled in with the speaker’s name); it will also interact with other users through comments through the interface.
👉 Click here to try Vocol AI verbatim service (portal)
Judging from our attempt to transcribe Elon Musk’s interview, the summary compiled by Vocol.ai is quite accurate. I won’t post the verbatim version. If you are interested, you can play it yourself (PS: It can be combined with the ones we have introduced) YouTube video download tutorial taking). From this interview video:
Vocol.ai excerpts this information in the abstract:
“ The conversation between Elon Musk and Bill Maher covered various topics such as the need for regulatory oversight for AI, the importance of free speech, and the dangers of censorship. Musk emphasized the need for a procurement process SOP to protect the public from potential dangers. The two also discussed Musk’s work with Twitter and his desire to ensure that the platform remains a trusted digital town square for a wide range of viewpoints. Despite their different backgrounds and areas of expertise, Maher and Musk found common ground in their shared concerns about the future of technology and its impact on society.
In a separate conversation, Speaker2 expressed their appreciation for Elon Musk and acknowledged that he had many choices and places he could go, but still chose to contribute to their project or organization. Speaker1 thanked Musk and ended the conversation, indicating that they would see him soon.
Elon Musk Discusses Procurement Process, Woke Mind Virus, and AI with Bill Maher
In this conversation between Elon Musk and Bill Maher, the two discuss a range of topics, including the need for a procurement process SOP, the dangers of the woke mind virus, and the potential risks of artificial intelligence. Musk highlights the importance of regulatory oversight for AI, arguing that it is necessary to protect the public from potential dangers. He also discusses the need for free speech and the dangers of censorship. Additionally, the conversation touches on Musk’s work with Twitter and his desire to ensure that the platform remains a trusted digital town square for a wide range of viewpoints. Despite their different backgrounds and areas of expertise, Maher and Musk find common ground in their shared concerns about the future of technology and its impact on society.
00:00:00->00:20:59Appreciation for Elon Musk
During a conversation between Speaker1 and Speaker2, Speaker2 expressed their appreciation for Elon Musk, stating that they couldn’t express how much they appreciated him. Speaker2 acknowledged that Elon Musk had many choices and places he could go, but still chose to contribute to their project or organization. Speaker1 then thanked Elon Musk and ended the conversation, indicating that they would see him soon.
00:20:55->00:21:04 ”
Although Vocol.ai currently does not provide one-click translation functionality (I believe it will in the future). However, if we simply throw it into ChatGPT for translation, we can easily get the Chinese version:
“埃隆·馬斯克和比爾·馬赫的對話涉及了各種話題，如AI的監管需要，言論自由的重要性以及審查的危害。馬斯克強調需要採購流程SOP來保護公眾免受潛在危險。兩人還討論了馬斯克在Twitter上的工作以及他希望確保該平台成為一個值得信賴的數字城市廣場，容納各種觀點。儘管背景和專業領域不同，馬赫和馬斯克在對未來技術的擔憂以及其對社會的影響方面找到了共同點。
在另一次對話中，發言者2表示對埃隆·馬斯克的欣賞，並承認他有許多選擇和去處，但仍然選擇為他們的項目或組織做出貢獻。發言者1感謝馬斯克，結束了對話，表示他們很快會再見面。
「埃隆·馬斯克與比爾·馬赫討論採購流程、覺醒思維病毒和AI」
在這次埃隆·馬斯克和比爾·馬赫之間的對話中，兩人討論了一系列話題，包括採購流程SOP的必要性、覺醒思維病毒的危險以及人工智能的潛在風險。馬斯克強調AI監管的重要性，主張有必要保護公眾免受潛在危險。他還討論了言論自由的必要性和審查的危險。此外，對話還觸及了馬斯克在Twitter上的工作以及他希望確保該平台成為一個值得信賴的數字城市廣場，容納各種觀點。儘管背景和專業領域不同，馬赫和馬斯克在對未來技術的擔憂以及其對社會的影響方面找到了共同點。對於伊隆·馬斯克的感激之情
在 Speaker1 和 Speaker2 之間的一次交談中，Speaker2 表達了對伊隆·馬斯克的感激之情，表示他們無法表達出自己有多麼欣賞他。Speaker2 承認伊隆·馬斯克有很多選擇和可以去的地方，但仍然選擇為他們的項目或組織做出貢獻。Speaker1 感謝了伊隆·馬斯克並結束了這次交談，表示他們很快就會再見。”
As for the QA video about Ada’s 100,000 subscriptions, it is summarized in Chinese:
“在這段影片中，電腦王阿達回答了觀眾的問題，分享了自己的經驗和看法。他提到了經營YouTube頻道和網站的不同之處，並說明瞭經營YouTube頻道需要花費更多的時間和精力。此外，他也分享了自己的學習經驗和熱情，並鼓勵觀眾要有興趣和願意學習。他還回答了觀眾的問題，包括自己的名字和經濟狀況等等。
電腦王阿達回答網友問題，分享經營心得
在這段影片中，電腦王阿達回答了一些網友對於他經營YouTube頻道和網站的問題。他分享了自己為什麼會想要經營YouTube頻道的原因，以及他如何開始拍影片的經驗。他也提到了經營網站和經營YouTube頻道的不同之處，並分享了自己在經營YouTube頻道方面的心得。此外，他還回答了一些關於自己的問題，包括他的名字和他的經濟狀況等等。整個影片充滿了電腦王阿達的個性和幽默感，讓觀眾感受到他的真誠和熱情。
00:00:00->00:10:37電腦王阿達的回答與展望
在這段影片中，電腦王阿達回答了一些觀眾的問題，包括自己學習3C知識的經歷、是否會拍臺南的VLOG、為什麼不加入B站等等。他提到，對於想要從事某一方面或某一路徑的人來說，最重要的是要有興趣並且願意學習，不要太計較薪水。此外，他也分享了自己對於頻道會員的想法，認為可以把廠商送的東西捐給慈善機構，並且預告了下半年可能會有一些採訪活動和旅遊題材的VLOG。最後，他呼籲觀眾們訂閱、按讚和開啟小鈴鐺，並期待下一支影片的到來。
00:09:10->00:12:44 ”
It can be seen that Vocol.ai not only summarizes the voice content in the video, but also knows how to add subtitles to the paragraphs. There is also a time period for marking the summary, which is quite considerate. Personally, I think it would be really useful to use this service in meetings or even for summarizing YouTube videos (generating subtitles is also good!) or audio interviews like the one we tried this time.
Vocol.ai also has many thoughtful options, such as a playback interface for listening to voice files by yourself, a verbatim search, and the ability to share verbatim analysis pages with other Vocol.ai users; the comment function can also mark or comment on certain voice clips. You can also directly click on the text in the verbatim area to simultaneously play the paragraph through the player (this is very convenient!).
That is to say, I believe that for this kind of application that uses AI to analyze meeting abstracts, everyone will pay special attention to the security part. According to Computer King Ada’s team, although Vocol.ai will use Azure GPT’s large language model for analysis, it will also add its own technology for processing in the middle. By default, there will be a mechanism for identification and encryption. Therefore, even if it is not only uploaded for sensitive information such as enterprises, they will be masked, that is, after personal information is removed, and then sent to GPT, which can provide a considerable degree of information protection.
However, I still want to remind everyone here that if there is really confidential information, you still need to consider whether to upload it to the cloud. However, as far as we know, in response to the demand for this kind of extremely confidential information, the Vocol.ai team will also have the possibility of local deployment in the future; for the interface part, the AI analysis function can also be turned off, and it will only be processed through the transcription function of Xidong intelligent speech recognition technology.
Oh, by the way, they seem to be preparing to launch a mobile app version in the third quarter, so you can look forward to it. As far as we know, the mobile browser can now support uploading recording files on the phone or even recording directly through the web interface – it will be temporarily stored on the local side of the web page and then uploaded to avoid disconnection. However, it is recommended to use a recording application that can avoid the screen automatically turning off to record meetings, which will be more stable, or look forward to their app version.
Personally, I am looking forward to a dedicated App version. I will let you know if there are any subsequent updates.
👉 Click here to try Vocol AI verbatim service (portal)
Further reading:
Source: KOCPC Chinese