An engineer’s report on a ‘fully automatic dictation diary’ that records his behavior 24 hours a day and writes it down with Whisper for a week – GIGAZINE

It was developed by the OpenAI artificial intelligence research grouphissis an artificial intelligence model that transcribes recordings with very high accuracy and is open source and free. In his blog, engineer Robert Dam reports test results of a system that allows you to leave “fully automatic dictation notes” by recording your actions on your smartphone every day and using Whisper to transcribe recordings. a plus.

I record myself on audio 24 x 7 and use artificial intelligence to process the information. Is this the future?
https://roberdam.com/en/wisper.html

Mr. Dam created fully automatic dictation notes because he believes that ‘if a smartphone came out with built-in storage beyond 1 terabyte, it would be possible to keep recording 24 hours a day, 365 days a year. It seems to start. In addition, OpenAI’s announcement that Whisper will be released to the public in September 2022 made a fully automated dictation diary a realistic idea, Dam said.

OpenAI Announces High-Performance Transcription AI ‘Whisper’, Supports Japanese and Can Transcribe Tongue Twisters and Lyrics with High Accuracy – GIGAZINE

Mr. Dam has a microphone made in China andsmall recorderAnd I always decided to record my actions out loud. At the time, it seems that by adding ‘Robert’ at the beginning of the content you want to talk about and ‘End Robert’ at the end, you can record all the content you’ve talked about. And he appears to have developed a system that processes all content recorded by Robert’s command at the end of the day with Whisper, converts it into a text file, and summarizes the content automatically.

On why Google Assistant doesn’t register the wake-word ‘OK Google’, Mr. Dam said: ‘I often wonder if you can do something interactively if you leave it as ‘OK Google’, or if it will display Google search results. “Because I don’t know”, “Because commands starting with ‘OK Google’ are saved in Google as audio files”, “Because there is a delay when commands are sent to Google”.

For example, say “Robert WEIGHT 62.8 end Robert” to record your weight for the day.

“Robert Sleep 7 hours 14 minutes (Sleep time 7 hours 14 minutes) End of Robert”

When you say “Robert LUNCH two toast with a fried egg (for lunch) end Robert”, it not only records the content of the meal but also automatically calculates the calories of what you ate using an external API. It seems that

Robert notes that the podcast is about Morgan Housell’s book The Psychology of MoneyThe psychology of money is the “wealth” mindset that will not worry about money for the rest of your lifeYou can write down your thoughts and ideas while driving by saying “Robert’s end.”

Then you can transcribe the content recorded with Whisper, convert everything into data, summarize it, and display it on the dashboard. As shown below, changes in body weight for one week, changes in sleeping time, calorie intake per day, total amount spent on gasoline and shopping, etc., are summarized in an easy-to-understand manner. .

In addition, it also has a “My Journal” function that automatically creates a diary that summarizes the day’s actions by showing the contents of the tweets tweeted for each hour of the day in chronological order.

“If everything you say gets recorded, it’s less paranoid. It’s about 100%,” said Dam. Also, not only the content of what is being said, but also 5W1H background such as ‘when, where, with whom and how’ is essential, so there was a limit to the amount of information that could be left just by recording and transcribing.

After trying the fully automated dictation system for a week, Mr. Dam said, ‘It’s magical to be able to bring back everything you did that day from small talk. ’” “Everything comes out of your mouth. By recording. All the exchanges and analyzing them, you can see things in a way that was impossible until now.” “The difference between utopia and dystopia is who has access to that information,” he said.