Hacker News Reader

Comment by chillacy | Hacker News Reader

I think you're right that chatbots need to be way more conversational, but I don't think having every company re-implement Cleverbot would be beneficial. Certainly Siri does this, but that gets old after awhile, and it's annoying when siri thinks you're looking for a joke or something clever when you just want information. I think the parallel to real life is when you call customer support, or order food at Wendy's. The conversation is usually extremely railroaded towards one specific subject (though within that domain, you can dive deep, like "are the pickles super sour or are they more like fresh pickles" that chatbots can't do yet)

View on HN

I think that's why Google's Assistant explicity went for "no personality, no banter".

I definitely agree with you on this. I think somewhere in the middle-ground is perfect, obviously.

I think I should be able to ask your chatbot how their day is going, but I don't think I should be able to shitpost with it for 3 hours.

The reason I bring up Chatbot is because they seemed to at least have the conversational part figured out.

Also, Siri is really not such a part of this conversation because it is honestly the only chatbot that I can think of that accomplishes a portion of what we're talking about here.

Siri can't even tell jokes, while Cortana is capable of telling jokes, singing, and much more.

These little things - even though programmed by some other person - can spice things up and make the bot look even more natural.

Go home Siri.

The other day, (French) Siri replied to "What are you doing this evening?" with "I'm playing hide-and-seek with Markov models".

This would have been a perfect example for the event domain I proposed to Apple. But they don't want it.

When I set a short timer the answer sometimes is something like:

"Make sure you don't overcook that egg" or

"The suspense is killing me".

Cortana's novelty functions are cute and make good advertising, but they're not really relevant to the quality of the service. In practice, you'll use them a couple of times, chuckle, and then never think about them again.

I never found a use case when Siri actually worked.

Failed cases : 1. Lookup Wikipedia during an amazing race 2. Play a specific song 3. Find an event in my calendar

2 works for me on Google's voice assistant. Setting alarms, getting directions, and voice dialing also work well. Nothing too exciting. Mostly useful when you don't have hands/eyes or don't want to get up and reach your phone. "Clap on/clap off" lights type of thing.

How is Google VA with background noise. I will renew my phone soon and am between the Pixel and the iPhone 7 (but I hate the absence of the headphone jack and don't want an adapter)

Hard to quantify. Some specific scenarios:

- It won't work in a loud bar

- It will almost always work in a car where you're the driver and there's a backseat conversation or the radio is on.

- It's fine with street/sidewalk noise.

- Not-terribly-loud coffee shops, stores, etc. are also usually just fine.

An anecdote: is that recently I was in a noisy restaurant with an open kitchen (think loud steam blasts and fryers and servers yelling out order numbers) and showing an older friend how far voice recognition had come and it successfully got most of our queries first try and if not, every time second try. This is with English language and asking it to play specific songs. We also have something of a strong accent compared to the san-fran types that presumably trained it (northern England)

Nowadays all those companies are reinventing the wheel: They make virtual assistants like Siri, Cortana, and Google Assistant. But really, they all do relatively the same things. These companies are just creating applied AI mainly for marketing and commercial reasons.

That's not how we'll be able to achieve real AI soon.

Reply on news.ycombinator.com