Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Amazon Polly – Lifelike Text-To-Speech

aws.amazon.com

Post reply on HN
Print view View original

91–100 of 106 posts

Re: Amazon Polly – Lifelike Text-To-Speech

#91

Post by nkozyra »

tyingq wrote: post #83
splike wrote: post #36

Earlier quoted context omitted.

I agree that the voice sounds pretty bad, but I think the 'lifelike' part comes from the intelligence part. I.E WA -> Washington, and 75F -> 75 Fahrenheit

> the 'lifelike' part comes from the intelligence part. I.E WA -> Washington, and 75F -> 75 Fahrenheit This would drive me nuts, as there's no easy way to predict what sort of transformations it would make. Basically, if it's not near perfect, it's arguably worse than no intelligence at all. With our PBX stuff, I know it's dumb, so I'm already accustomed to giving it something I know it will read correctly, like "dom…

> This would drive me nuts, as there's no easy way to predict what sort of transformations it would make.

I believe that's what the Lexicons are for within the service.

Re: Amazon Polly – Lifelike Text-To-Speech

#92

Post by dom0 »

TazeTSchnitzel wrote: post #16

Could we drop “Lifelike” from the title? It's marketing puffery (and untrue, it sounded very robotic to me to the point I was convinced I'd heard the wrong audio file). The actual story is that it's Amazon's text-to-speech-as-a-service offering, not that it's particularly innovative in terms of sounding good.

Joel sounds like Microsoft Sam cleaned his blocked nose. shudders

Re: Amazon Polly – Lifelike Text-To-Speech

#93

Post by tyingq »

nkozyra wrote: post #91
tyingq wrote: post #83

Earlier quoted context omitted.

> the 'lifelike' part comes from the intelligence part. I.E WA -> Washington, and 75F -> 75 Fahrenheit This would drive me nuts, as there's no easy way to predict what sort of transformations it would make. Basically, if it's not near perfect, it's arguably worse than no intelligence at all. With our PBX stuff, I know it's dumb, so I'm already accustomed to giving it something I know it will read correctly, like "dom…

> This would drive me nuts, as there's no easy way to predict what sort of transformations it would make. I believe that's what the Lexicons are for within the service.

If all of the auto transformations are defined Lexicons, that can be edited/removed, that would be fine.

I read the post I was replying to as some sort of undefined, built-in, set of intelligent transformations.

For example, 75F-> 75 Fahrenheit would require that you make a Lexicon entry for every value...1F, 2F, 3F, etc. That's not really intelligent :)

Re: Amazon Polly – Lifelike Text-To-Speech

#94

Post by visarga »

dharma1 wrote: post #50
IshKebab wrote: post #43

Earlier quoted context omitted.

Sooo cynical. This is way better than most TTS systems, including Google's. The only one I've ever heard that is better is DeepWave, and that suffers from lots of background noise.

Do you mean WaveNet? The noisiness is because it uses 8bit (mu-law) audio. 8bit is because WaveNet is computationally extremely expensive. It's pretty much proof of concept at this stage, not production ready TTS.

There is a faster variant on GitHub, called, of course, Faster-WaveNet.

Re: Amazon Polly – Lifelike Text-To-Speech

#95

Post by visarga »

jeena wrote: post #49

I really like text2speech because I really like the computer to read articles to me while I'm doing something else, but I won't use any internet service for that just to get a little bit more quality in comperison to Apples "Alex" or MeryTTS http://mary.dfki.de/

Me too. I usually play Reddit and HN on Alex voice. It's the best instantly accessible voice on Mac, even though it is quite old.

Re: Amazon Polly – Lifelike Text-To-Speech

#96

Post by awqrre »

TazeTSchnitzel wrote: post #16

Could we drop “Lifelike” from the title? It's marketing puffery (and untrue, it sounded very robotic to me to the point I was convinced I'd heard the wrong audio file). The actual story is that it's Amazon's text-to-speech-as-a-service offering, not that it's particularly innovative in terms of sounding good.

It's more like the opposite of being innovative (unimaginative?)... They bring a product that could previously achieve that locally to a remote server and the quality is not any better.... and they want to charge you by the character... no thanks Amazon... hopefully it's free with Prime.

Re: Amazon Polly – Lifelike Text-To-Speech

#97

Post by justusthane »

j2kun wrote: post #22
jicks wrote: post #7

Earlier quoted context omitted.

I agree, it doesn't sound "lifelike" at all to me. Compared to WaveNet[1] it's day and night. [1] https://deepmind.com/blog/wavenet-generative-model-raw-audio...

The "babbling" samples here are really fun. Would make for a great "Sims" language.

I actually found that pretty unsettling to listen to, and verging on creepy. I felt like I was listening to a computer trying to pretend to be human.

Re: Amazon Polly – Lifelike Text-To-Speech

#99

Post by mgamache »

mtgx wrote: post #67
mgamache wrote: post #60

There's a lot of negative reaction to Polly. Besides WaveNet (which is not available as an API-- and really slow), what are better alternatives? Or is the reaction just to Amazon's marketing speak using "lifelike" and "deep learning"? I find most of the voices to be okay, but Joanna is pretty good and better then anything I've heard besides WaveNet. Also, with Amazon's FPGA investment how long before it implements Wa…

Why would Google use FPGAs over its own (faster) TPUs? Chances are Google either doesn't have enough TPUs (it may be waiting on the next generation to go massive scale with them), or Wavenet is still too computationally intensive even for TPUs.

I am sure Google is looking at WaveNet & TPUs, I was just thinking Amazon would look at using its new massive FPGAs to implement WaveNet. I have no idea if the FPGAs would be fast enough.
Post reply on HN