How to make anki enable voice interaction or AI voice interaction? Greatly enhance immersion and free your hands
Anki enables voice interaction to greatly enhance immersion and make learning easier. It also frees up your hands and eyes in revision scenarios where you don’t need to look at the screen.
1, “voice-activated buttons”: this is the simplest voice interaction I can think of, the need to click on the button to achieve voice interaction, such as “show the answer” and "show the answer after a few options "This way, we don’t have to press the keyboard or touch the screen of the mobile phone frequently to complete the anki button clicking.
2、“Automatic reading of the card”: this is a more advanced way of interaction, in the “voice-activated buttons” on the basis of “automatic reading of the card content”. This makes it easy to review while exercising, walking, taking a bath, or resting with eyes closed. (“Automatically read out the card” I know can be achieved through AwesomeTTS, but using it to make the card still need multiple steps is not convenient, and also lack of voice interaction )
3, “AI voice interaction”: using GPT4, Gemini Ultra and other high-quality large models (or local open-source free offline small models), so that the AI with AIKI to achieve better and more natural voice interaction. This is like a human-to-human dialogue, which can greatly improve the sense of immersion and free both hands and eyes.
Is there any big brother has any relatively easy to operate method to achieve any one of the above 1, 2, 3, or can be written as a plug-in for us to use. I don’t know how to program myself and I’ve tried a lot of ways to achieve this.
1 Like
Absence of voice control feature
The majority of the team of AnkiDroid maintainers have agreed that there will be no “voice commands” feature in AnkiDroid. AnkiDroid as an organization will not help implementing such a feature and strongly prefer that all AnkiDroid contributors abstain too.
This decision is not a technical one but an ethical one. Even in 2011, the first time this feature was requested, many users told us that voice-recognition would allow them to review while driving. AnkiDroid maintainers do not want to contribute to distracting users while they drive. Distraction is a major cause of vehicle crashes. While we can’t avoid drivers being distracted by other apps, at least we can avoid providing an additional distraction. We want to emphasize that, since vehicle crashes involve not only the user, but also any by-stander or passenger, we can’t even ethically decide to let the user assess the risk level to themselves, as this can and will have impact on third parties.
As a flash-card app, we did not foresee the need to deal with ethical questions in our software, and had not previously discussed an ethical code. For the sake of simplicity, we will thus simply refer to ACM’s one. Rule 1.2 states “Avoid harm”. In particular “especially when those consequences are significant and unjust”. Furthermore, the code of ethics mentions all people, and not only the user of the computer.
There is, to this day, no way to disable a feature when someone is driving; even with GPS coordinates we can’t distinguish between a passenger, a driver, or maybe a public-transit passenger. Even if our term-of-service require not to use the feature while driving - which may already be against the law in many places anyway - it is a certainty that, with millions of users, many of them being stressed students with little free time, those restrictions will not be followed and uninvolved bystanders will still be harmed. It is thus safer today to just avoid this feature altogether.
We recognize that this feature could have valid use-cases, reviewing hand-free in less dangerous situations. In particular we recognize that we are rejecting an accessibility feature. While it’s not a happy decision, we believe that the trade-off between avoiding road accidents and an accessibility feature is very strongly in favor of avoiding accidents.
Absence of voice control feature · ankidroid/Anki-Android Wiki · GitHub
While this only is about AnkiDroid, the same argument can be used to avoid such feature in AnkiMobile
If both plataforms are not using the voice interaction, then why would anki add it
Extra: AI plugins
2 Likes
That AnkiDroid voice control discussion is interesting.
Previously I tried voice control by customizing it myself (without driving), but the mobile device heats up and drains the battery, so it wasn’t very practical.
I think just studying the material, like audiobooks will be good. Answering cards requires concentration, so multitasking is difficult to begin with.
1 Like