The State of Text-to-Speech Technology

If you've been enjoying the many text-to-speech voices available inside our Read Aloud browser extension, you may have gotten the false impression that these voices are free.  They're not.  In fact, they're quite expensive.  In the table below, we explain how we're able to provide some of these voices for free, and why we're charging for others.

Voices How much it costs How we provide it
Google voices These voices are built into the Chrome browser and are free. You can use them as much as you want.
Google Translate voices These voices are not free. In fact, Google never intended to make them available to 3rd party applications. But thanks to open-source software, we're able to bring them to you. Google might decide to cut off access to these voices in the future, however.
Google Standard and Google Wavenet voices These voices are not free. Standard voices cost $4 per million characters, Wavenet voices cost $16 per million characters.

(Google has a generous free tier if you have a Google Cloud Platform account.  Read more below.)
We're able to bring you these voices for free courtesy of Google's promotion. We don't know when this promotion will end. So these voices could go away at any time.
IBM Watson voices These voices are not free. Watson voices cost $20 per million characters. We're able to bring you these voices for free courtesy of IBM's promotion. We also don't know when this promotion will end. So these voices could go away at any time.
Amazon Polly voices These voices are not free. Polly voices cost $4 per million characters.

(Amazon has a generous free tier if you have a Amazon Web Services account.  Read more below.)
There is no promotion for Amazon voices. But thanks to open-source software, we're able to serve them to you from our cloud servers at a significant discount. You can access them via the Premium Voices in-app purchase.
Microsoft Desktop voices These voices are free if you have a Windows computer. Follow these instructions. We're serving Microsoft Desktop voices to you from our cloud servers. And to cover operational costs, we have to put them behind the paywall. You can access them via the Premium Voices in-app purchase.


As you can tell, these prices are not even close to being affordable. And you can tell that we're doing our best to bring these voices to you free of charge. Only for Amazon and Microsoft voices, we have to put them behind a paywall to cover our operational costs.

Custom Voices


We have implemented a "Custom Voices" feature, allowing you to use your own Google Cloud Platform (GCP) or Amazon Web Services (AWS) account to pay for your own usage of the Google Wavenet and Amazon Polly voices.  Most likely you will get them for free because of the generous free tiers offered by Google and Amazon.  However, the setup process is fairly involved and requires a credit card.  If there is enough demand, I will make some instructional videos to help you through it.  Leave a comment below.

If you're a school or organization, we also recommend using your own GCP/AWS account so that you have total control over the availability and cost of your usage.  With the generous free tiers, it's unlikely you'll actually incur charges.  (Note, however, that the AWS free tier expires after 12 months)

130 comments:

  1. I'm just curious, when did you start the premium voice fee system? I've been using your extention for awhile and this was the first I saw of it, and perhaps I just had gotten to the end of a trial, but I'm not sure. I'd love to know which one I was using before the switch, but I fully intend to pay to continue using it as it's been an awesome improvement in my life.

    ReplyDelete
    Replies
    1. Hi Stephanie, it started about a week ago. Amazon and Microsoft voices are now fee-based. You can test out those voices from the purchase page to see which voice you were using earlier. Please contact support@lsdsoftware.com if you need more help.

      Delete
  2. This is ridiculous. I don't want voices streamed from a server I want them generated locally. I don't want to sign into google to a plugin. I don't want your servers having access to everything I read. This is immoral and you should feel bad. I had an old read aloud plugin that was working until recently when chrome started blocking it because it had not been updated in a while. Freaking planned obsolescence enforced by google. Your blatant cash grab is noted and unappreciated. I will find a suitable alternative or create one myself, as software projects go this one is not that difficult.

    ReplyDelete
    Replies
    1. If you like, you can purchase individual IVONA voices from harposoftware.com. They cost $45 per voice.

      The old Read Aloud plugin was streaming from the server too, we just haven't blocked it yet. But soon we'll have to block the old plugins, because our bank account is almost empty! Unless we collect some fees we can't continue to pay for those servers!

      If nobody wants to pay, great, we'll just shut down those servers and remove the Amazon & Microsoft voices. But we can't have those servers and not pay the hosting companies! It's like renting an apartment. Every month you have to pay rent, or you get kicked out.

      Delete
    2. Hi Frank, that's not exactly fair. The voices are extremely sophisticated AI software, at least the wavenet and others. They use a symbols approach to streaming so you aren't using much data. This extension is fantastic for its simplicity of use and more than that, the fact that they have these voices available (at least for now) and the wavenet ones from google as of this message are still completely free in this extension (and by far the best IMO). Before I found this extension, I had to get the appropriate account with google then follow the extremely complex instructions to convert some text to speech using their API and python script. I don't know of any other way to get access to the google wavenet voices than this extension or writing your own code in python. Actually I think they let you demo it for text you write in a text box on their website but not in any kind of end user application like this browser extension.

      These newer sophisticated AI type voices are not free from the people who created them, well they have free trials and promotions but essentially they are a premium paid product that has been developed by google and others aimed at software and app developers. There are plenty of products out there using microsoft or other voices that don't need to stream anything etc but they don't sound as good. It's really difficult to make something sound convincingly like a human, it's only very recently that it's become accessible to us mere mortals so maybe in time there will be standalone ones released for free by the likes of google but for now we are using their technology developed by them and it's up to them how and who gets to use it and how much it will cost. If/when the google promotion ends, the voices will have to go away in this extension unless there is a paid option used because the developer of this extension will have to pay for access to the voices and they need to cover those costs.

      I actually stopped by to pay complement to the developer and thanks for the access to the wavenet voices while it lasts. Thanks a bunch, really nice and simple UI with everything I could want and no unnecessary bloat :) Nicely done!

      Delete
    3. Thanks for the compliments! You're spot on.

      Delete
    4. Frank, your summary of the situation is laughably inaccurate. You can use on-machine text-to-speech software, but they're typically not as sophisticated as what cloud computing services can provide, due to the processing power required to deliver the service.

      Delete
    5. Google WaveNet is currently state-of-the-art deep neural networks (DNNs) model. The model is fairly large and I think it made perfect sense that google deploy and running them on their cloud.

      @Holy Cow: your extension rock! I love it.

      @Alex West: been there done that =))

      Delete
  3. And why pray tell cant you simply use the web speech api like god intended when he created the internet? https://w3c.github.io/speech-api/speechapi.html

    ReplyDelete
    Replies
    1. We Are using the Web Speech API, which makes available only the browser's built-in voices. In Chrome, the built-in voices are your "Google XXX" voices, which are indeed generated locally and free for unlimited use.

      Delete
  4. Can't you add Ad's like everyone else does, and offer the app w/out Ad's to those who will pay for premium service ?

    ~Just Sayin~

    ReplyDelete
    Replies
    1. Yes, we've considered this. However, popular ad services don't support Chrome extension because there's no context to serve the ads. Less popular ad services would do things such as injecting ads into the web page you're on (the webpage is the context), but this is shady and is now forbidden by Google after the incident where millions of Facebook account were hacked via Chrome extensions.

      Delete
  5. Do you know what Google TTS's en-us-x-sfg#male_1-local voice is under your list of supported voices? Thanks.

    ReplyDelete
    Replies
    1. I don't recognize what voice that string corresponds to, sorry.

      Delete
  6. muchas gracias por las voces en español

    ReplyDelete
  7. Hi there, just wanted to say thanks, I have been using this so my eyes don't water so much with all the reading I have to do for university. I paid the $4 for the wavenet voices but they seem to have been removed from the list now. I'm guessing it's the financial reasons mentioned above. Anyway, I'm happy to use the custom voices method, I'm just a total novice and not sure how to 'secure' my API key and soforth. So a video tutorial would be much appreciated if you get the time :)

    ReplyDelete
    Replies
    1. Hi, yes I'll try to get to that sometimes. In the mean time, you can try this video: https://www.youtube.com/watch?v=ZUHqMn6NacY .Remember to enable the Cloud Text-to-Speech API, not the Speech (Recognition) API mentioned in the video. If you need help, please email support@lsdsoftware.com directly for quicker response!

      Delete
  8. Okay so first things first, I want to let you know your extension is awesome! Probably one of my most used extensions.

    I wanted to make a few suggestions that I believe would (in my opinion) make it even better!

    So follow me here...
    When I use the "wavenet TTS api" in my cloud console, I can export the read text to an mp3 format. I know its already possible to upload a pdf to read but if you could also output the read text into an mp3 format that would be very useful(See Python Example).

    See Example: https://codepen.io/dynamiqai/full/qgEqLR

    ReplyDelete
    Replies
    1. Glad you find it useful! For the MP3 download it's somewhat complicated because when you read an article, the audio is synthesized paragraph by paragraph as you read. This is to avoid synthesizing the entire article in case you stop reading early (to save on bandwidth and cost since characters are so expensive). If you read aloud an article, and then synthesize the whole article as MP3, we may end up synthesizing the text twice.

      Unless we use the server to cache the paragraph fragments and concatenate them to create the final MP3. In the case of Custom voices, however, the synthesized speech are downloaded directly from Google Cloud to your computer, and never passes by our server (neither does your API key-for security). So we can't do the caching and concatenating.

      I'm still thinking of how to make this feature completely separate from the main Read Aloud function, so that for a specific article you can choose at the beginning whether to read it aloud in the browser, or to synthesize it as MP3. One or the other, and not both.

      Delete
    2. Well that makes perfect sense to me, if you do end up creating that feature I would love to provide input. Also if you'd like any side development projects, I know of a few businesses that may be interested in working with you. Would it be possible to get in touch with you? I can be reached @ tnm@dynamiq.ai

      Delete
    3. Also on another note, what happened to the pitch settings in the latest version?

      Delete
    4. Ah, the pitch setting doesn't work for any of the cloud-based voices. It'll appear only when you select one of the local voices like the Google US English voice.

      Delete
  9. This is the best TTS I've ever used simply because of the wavenet voices and how easy it is to use. Hopefully they won't become unavailable any time soon, but just in case they do it would be great if you could set up a tutorial on how to set up wavenet on Google Cloud

    ReplyDelete
    Replies
    1. Yes, will do it as soon as I find some time. Thanks

      Delete
  10. Missing host permission for the tab - Don't know what to do here.

    ReplyDelete
    Replies
    1. Please use the in-app report-issue feature so we have some context to diagnose the problem. Usually this happens if you click the Read Aloud button while on a special browser page that doesn't start with "https://". Or the focus isn't on the page you want to read aloud.

      Delete
  11. If I buy a harposoftware voice, how do I use that with Read Aloud? Their site says the voice will come with a piece of software, but I rather enjoy your Chrome plugin and would rather use that. Thanks.

    ReplyDelete
    Replies
    1. Hi, after you install the Harpo voice, if it is an IVONA voice for example, there'll be a new entry automatically appear in the voice list that looks like "IVONA 2 XXX OEM". You can select that voice to use it.

      Delete
  12. You've got a great text-to speech implementation here. I paid because I think its worth it.

    ReplyDelete
    Replies
    1. Thank you. Well, we're making only enough to pay the cost. Currently it's about $100/month enough to cover the small number of users using the premium voices. I'm working pro bono, pretty much.

      Delete
  13. Is it possible to add the filipino Google translate voice in to the list of supported voices? If so, the addition would be much appreciated. Thank you!

    ReplyDelete
    Replies
    1. I have deployed version 1.5.1 which should support Filipino language. Thanks!

      Delete
  14. Hi,
    Is there a way to record the audio and maybe download it?

    Cheers

    ReplyDelete
    Replies
    1. Hi, sorry the extension doesn't have that feature at this time. Please try https://ttstool.com instead

      Delete
  15. This is one of the most powerful and useful tools for implementing cloud based TTS technology. Than you so much! I am using my Google API key to access google wavenet voices and then chrome extension to read aloud my ebooks on play store. I use google translation -english voice as a backup. Everything works wonderfully, great job on this!

    ReplyDelete
  16. Hi, I've been using your service for a while now and I'm really impressed with it, especially the wavenet voices. I purchased the amazon voices but they don't sound as good as when I'm using my Amazon Web Services Polly API. It sounds like it is synthesizing one single word at a time rather than sentences so the pronunciation is strange. Do you use AWS for the voice or a separate vendor? Because when I used different vendors it sounded a lot worse than when I used AWS Polly. Also, my favourite Amazon Polly voice is Matthew (US English) but it isn't available on this app, is there a reason why?

    ReplyDelete
    Replies
    1. Hello, we use an outside vendor (instead of AWS) to provide the Amazon voices, because AWS would simply be too costly. The outside vendors use older versions of the speech engine that doesn't have machine-learning technology, that's why it doesn't sound as good. And the Matthew voice is a new voice available only on AWS.

      Delete
    2. Thank you for the reply. It's a shame Matthew isn't available but it's a great app nonetheless.

      Delete
    3. This comment has been removed by the author.

      Delete
    4. Hey there,
      Just wanted to mention that AmazonPolly US English (Matthew) is available if you have an AWS account and add your AWS Credentials to Enable Custom Voices. (it's the last option at the bottom of the voices List) or try...
      AWS Credentials to enable Amazon Polly voices
      chrome-extension://hdhinadidafjejdhmfkjgnolgimiaplp/custom-voices.html

      PS.
      IBM-Watson American English (Michael) & (MichaelV2) are also Awesome!!

      Delete
  17. Replies
    1. Could you elaborate what happened when you click Read Aloud? Do you get an error, or does it appear to read but no sound is heard, or is it stuck spinning?

      Delete
  18. Would it be possible to add (Pitch) as one of the options for wavenet voices?

    ReplyDelete
    Replies
    1. Are you referring to the Custom voices where you use your own GCP account? Because the free Wavenet voices are no longer available; Google has rescinded their free promo.

      Delete
    2. Are you referring to the Custom voices where you use your own GCP account? Because the free Wavenet voices are no longer available; Google has rescinded their free promo.

      Delete
  19. hello uh
    what happened to google standard voices
    they're gone for a long time now, and i can't even buy them ><
    in love with colleen voice ><

    ReplyDelete
    Replies
    1. Halo, Google has finally blocked us from using those voices for free. After all, they're really good and cost money. What you can do is sign up for the Google Cloud Platform account and use the Custom Voices feature to pay for it yourself--though you won't actually pay anything because of the free tier. Unfortunately, as it's intended for developers, it's somewhat complicated to setup.

      Delete
    2. @Holy Cow are the steps to do this what I see at https://cloud.google.com/text-to-speech/docs/quickstart-protocol if I do steps 1 to 4.a This seems a better explanation than what I read on https://cloud.google.com/free

      I don't see it listed in free tier intro page https://cloud.google.com/free . Does this api key use the $300 free credit ?

      Delete
    3. Yes! Actually you only need to complete the first 3 steps. Then in step 4a, instead of creating a Service Account Key, create an API key to use in Read Aloud.

      The free tier for this service is explained on this page: https://cloud.google.com/text-to-speech/pricing

      Delete
  20. Frank, I am with you buddy, this aggravates me to no end but I see this as a challenge. I am one of the reigning Bargain Shopping Queens, or BSQ for short ;) I know there is a way to get around this, I just have to find the right path.
    My field is Information Systems - Data Analysis, kissing-cousins to IT people so I have a great deal of respect and appreciation when it comes to AI & developing a good TTS program - SALUTE. I can only speak in nouns, adjectives, verbs, etc. Computers speak in ones and zeros. For that reason I am thankful there are people out there that are blessed with being able to translate English to Binary.
    One last thing Frank, I thought for sure it was Al Gore that invented the Internet.
    :D

    ReplyDelete
    Replies
    1. I wrote a tamper-monkey script that uses the web speech api (on windows it also has all the Microsoft voices) I have it hard coded to my speed and voice preferences. It's kinda buggy (I'm not a web developer) and it kinda some times fights with some websites that already catch hotkeys (a relatively small percentage of them), but it's pretty easy to disable when needed. I'm sure it would be trivial for someone with more skill to polish something similar up that wont have bugs but I wrote it in 30 minutes and it's good enough for me so there it stands, feel free to use it bugs and all.

      I hereby relinquish copyright claims on the following work and issue it to the public domain for the benefit of humanity. Feel free to use it or incorporate it in other works, attribution not required.

      //Paste this code in as a new tampermonkey script
      //To use highlight text on a website and press ctrl + alt + z

      // ==UserScript==
      // @name New Userscript
      // @namespace http://tampermonkey.net/
      // @version 0.1
      // @description try to take over the world!
      // @author You
      // @match *://*/*
      // @grant none
      // @require http://code.jquery.com/jquery-latest.js
      // ==/UserScript==

      (function() {
      'use strict';

      // Your code here...

      document.addEventListener ("keydown", function (zEvent) {
      if (zEvent.ctrlKey && zEvent.altKey && zEvent.key === "z") { // case sensitive
      // DO YOUR STUFF HERE

      function getSelectionText() {
      var text = "";
      if (window.getSelection) {
      text = window.getSelection().toString();
      } else if (document.selection && document.selection.type != "Control") {
      text = document.selection.createRange().text;
      }
      return text;
      }

      var msg = new SpeechSynthesisUtterance();
      var voices = window.speechSynthesis.getVoices();
      msg.voice = voices[10]; // Note: some voices don't support altering params
      msg.voiceURI = 'native';
      msg.volume = 1; // 0 to 1
      msg.rate = 8; // 0.1 to 10
      //msg.pitch = 2; //0 to 2
      msg.text = getSelectionText();
      msg.lang = 'en-US';

      msg.onend = function(e) {
      console.log('Finished in ' + event.elapsedTime + ' seconds.');
      };

      speechSynthesis.speak(msg);

      }
      } );
      }
      )();

      Delete
  21. I would love to contribute, even though I am low income. Is there any way to buy you a coffee? Or support you in some other small way?

    ReplyDelete
    Replies
    1. Thank you for your good will! There should be a buy-me-coffee link on https://www.lsdsoftware.com

      Delete
    2. Yes, definitely fair coffee :)

      Delete
  22. This comment has been removed by the author.

    ReplyDelete
  23. Excellent job with this extension.
    I'll put the link of this blog to the google chrome extension page.
    Again, great work, please keep on doing it!

    ReplyDelete
    Replies
    1. Thanks for your compliments and evangelism!

      Delete
  24. Just that I wish you had simply made it clear from the top when I installed this extension in Chrome that there was only 1 voice choice available. I went through all the USA voices before finding that there were never any other choices available uintiol I paid! he voice and all were unavailable until I paid$$$$$$$$$$$$ UNFAIR AD PRACTICE!!!

    ReplyDelete
  25. I loveeeee this extension. I have been looking for audiobooks online but they don't have the book I wanted so this is a game changer!!!!!! There are new voices available for OpenFPT Vietnamese that sounds much more natural but I don't know if you are able to add that in the system. But great job guys! Definitely makes my life so much easier!

    ReplyDelete
    Replies
    1. Hi, thanks for feedback! OpenFPT voices are available in the voice list, towards the bottom. It may sometimes not work, however.

      Delete
  26. I added my credit card to GCP and enable tts. However, Google still asks me to prove not-robot. Could you provide setup youtube video?
    Thank you for awesome extension!

    ReplyDelete
    Replies
    1. Alo, you need to create an API key then enter it in the Custom Voices section. Go to Options, in the Voices drop-down, scroll to the bottom and choose "Enable Custom Voices"

      Delete
  27. Love your plug-in! Life changer! Is there anyway to add a d"ownload to MP3 feature"? Or do you know of any software that can do this? I'm using the free tts reader Balabolka that has this function, but it does not allow you to use all the voices like your plugin. It takes a long time, but it will allow you to create an MP3. http://www.cross-plus-a.com/balabolka.htm

    ReplyDelete
    Replies
    1. Have you tried https://ttstool.com ? It may be what you looking for.

      Delete
  28. I keep trying to use the WaveNet voices and it works for a little, then it tells me I no longer have permission. What can I do to fix this. I've even tried paying for a package and style no luck. Please help ASAP.

    ReplyDelete
    Replies
    1. Hello, best option is to sign up for your own Google Cloud Platform account, create an API key, then enter the API key into the "Enable Custom Voices" section of Read Aloud. Then you won't be asked for those permissions.

      Delete
  29. I haven't paid and i also have this female voice, and found a few others that worked with the TTST.
    https://chrome.google.com/webstore/detail/us-english-female-text-to/pkidpnnapnfgjhfhkpmjpbckkbaodldb

    ReplyDelete
  30. Thanks for a great extension, I use it daily.
    There is a new version of IBM Watson V3,but in the extention only v1 and v2.

    ReplyDelete
  31. Hi, this is JJLin from aliprice.com, i like your extension, it's useful.
    We are getting more users our Chrome extension(AliPrice Price Tracker). Maybe you can suggest it to your users who really need it.
    It's not usual ads and i will tell you more if you interesting.
    email: lin@aliprice.com
    And you will get paid.

    ReplyDelete
  32. First and foremost, from the bottom of my heart, thank you for this extension! My question pertains to CereProc Text-to-Speech (https://www.cereproc.com/). I was just curious if Read Aloud would ever consider including their voices? Ivona (now Amazon Polly) and CereProc are, in my opinion, the most realistic sounding. CereProc was hired to recreate an authentic President John F. Kennedy voice based on previous speeches that he had given so that he could finally deliver the speech he was meant to deliver in Dallas 55 years ago, had he not been assassinated. Here is a link to the full recording: https://www.cereproc.com/en/testpage. Personally, I have noticed that their voices are characterful, engaging, and emotional but don't take my word for it. Simply listen to some samples here: https://www.cereproc.com/en/products/voices.
    They even have some celebrity voices!

    I am obviously passionate about the quality of their voices because I have been experimenting with text-to-speech since the late 90's and I love how far the technology has come in the past 2 decades. Read Aloud has increased my reading comprehension and truly improved my study habits. I know the voices I mentioned will make the extension even better so I just want to bring it to your attention, that's all. Honestly, I use the extension everyday so I'm just making a friendly suggestion so that it can remain the best of the best. It's improving the quality of life for my wife, twins and I, so thank you immensely. Please keep up the great work. You really are making a difference and I want you to know it! :)

    ReplyDelete
    Replies
    1. Thank you for the wonderful feedback! I'm glad it has been useful for you. We'll see what can be done with CereProc. I created an issue to track this: https://github.com/ken107/read-aloud/issues/120

      Delete
  33. cmn-CN-Wavenet-A, cmn-CN-Wavenet-B, cmn-CN-Wavenet-C, cmn-CN-Wavenet-D are unavailable to choose from even the Chinese language is enabled. GoogleTranslate Chinese is available though. But not the Wavenet ones. How can I enable the Wavenet ones?

    ReplyDelete
    Replies
    1. Looks like I need to add them. Will do for the next release. Thanks for reminding!

      Delete
  34. Hello,
    Can someone post a link to step-by-step guide on how to link your Google account to enable wave-net voices? I have done that last year on my old laptop, now cant find it :/

    ReplyDelete
    Replies
    1. Hello, the steps are as follows:
      1) Sign up for a Google Cloud Platform account and create a Project
      2) Go to API & Services - Library, find and enable the Text-to-Speech API for the project
      3) Go to API & Services - Credentials, and create an API key
      4) Enter the API key into the Custom Voices section of Read Aloud

      Delete
    2. At step 4, when enter my API I get the following error.

      Test failed: failed to fetch https://texttospeech.googleapis.com/v1beta1/voices?key=AIzaSyDahJzBa94h_NwVS-qO0Rtj9k2PKcHimXM

      I have since deleted the API so I could publish it here.

      Delete
    3. Could you recreate your API key, then open that URL in the browser and tell us what you see?

      Delete
    4. Here's what I got. I'll post three replies covering the tree tabs I see. This first one is JSON.

      error
      code 403
      message "Cloud Text-to-Speech API has not been used in project 953921189679 before or it is disabled. Enable it by visiting https://console.developers.google.com/apis/api/texttospeech.googleapis.com/overview?project=953921189679 then retry. If you enabled this API recently, wait a few minutes for the action to propagate to our systems and retry."
      status "PERMISSION_DENIED"
      details
      0
      @type "type.googleapis.com/google.rpc.Help"
      links
      0
      description "Google developers console API activation"
      url "https://console.developers.google.com/apis/api/texttospeech.googleapis.com/overview?project=953921189679"

      Delete
    5. This is the raw data tab.

      {
      "error": {
      "code": 403,
      "message": "Cloud Text-to-Speech API has not been used in project 953921189679 before or it is disabled. Enable it by visiting https://console.developers.google.com/apis/api/texttospeech.googleapis.com/overview?project=953921189679 then retry. If you enabled this API recently, wait a few minutes for the action to propagate to our systems and retry.",
      "status": "PERMISSION_DENIED",
      "details": [
      {
      "@type": "type.googleapis.com/google.rpc.Help",
      "links": [
      {
      "description": "Google developers console API activation",
      "url": "https://console.developers.google.com/apis/api/texttospeech.googleapis.com/overview?project=953921189679"
      }
      ]
      }
      ]
      }
      }

      Delete
    6. This is the headers tab.

      X-Firefox-Spdy h2
      alt-svc h3-29=":443"; ma=2592000,h3-27=":443"; ma=2592000,h3-25=":443"; ma=2592000,h3-T050=":443"; ma=2592000,h3-Q050=":443"; ma=2592000,h3-Q046=":443"; ma=2592000,h3-Q043=":443"; ma=2592000,quic=":443"; ma=2592000; v="46,43"
      cache-control private
      content-encoding gzip
      content-length 404
      content-type application/json; charset=UTF-8
      date Thu, 09 Jul 2020 07:53:00 GMT
      server ESF
      vary Origin, X-Origin, Referer
      x-content-type-options nosniff
      x-frame-options SAMEORIGIN
      x-xss-protection 0
      Accept text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8
      Accept-Encoding gzip, deflate, br
      Accept-Language en-US,en;q=0.5
      Connection keep-alive
      DNT 1
      Host texttospeech.googleapis.com
      Upgrade-Insecure-Requests 1
      User-Agent Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Firefox/78.0

      Delete
    7. The response headers was from start to x-xss-protection 0 Then the request headers followed.

      Delete
    8. AW HA! 😲 I followed the console.developers.google.com URL on the JASON tab and it took me to a page talking about the the text to speech API with a button that said enable. I clicked the enable button and then went back to the Read Aloud API entry page and entered the same API again and it excepted it that time. 😁 I tried the same API it a few times prior to that so that enable button is most likely what got it working.

      Delete
  35. Wavenet suddenly became unavailable. It was working until just now. Why?

    ReplyDelete
    Replies
    1. Alo, could you tell me if you're using your own Google Cloud Platform API key? The Wavenet voices have some issue, I'll debug and get it back on asap.

      Delete
    2. Yes I am using my own API key which should work. Thanks in advance.

      Delete
    3. There news to use with own API key???

      Delete
    4. Yes, it should all be working now. Thanks

      Delete
    5. it works for me but i still have to "get permission" and confirm i'm not a bot every time i use it, even though i'm using my own API. is everyone using wavenate like that as well or am i missing something here?

      Delete
    6. Hi Moe, if you've entered your API key, then you should not get the "permission" prompt any more. Please go to the "Enable Custom Voices" page and check that your API key has been entered into the box in the section for Google Wavenet.

      Delete
  36. Nice article, Which you have shared here about the Text-to-Speech Technology. Your article is very informative and useful to know more about the Different types of Text-to-Speech converters. If anyone looking for the text to speech mp3 converter free download, voicemaker is the best choice.

    ReplyDelete
  37. There are an awful lot of Chinese voices listed -- most of them nonexistent. It's nice that you can waylay me into ten minutes of finding out why -- but it would have been nicer if you'd just removed them and saved me the time.
    Might you, btw, label the voices as male and female?

    ReplyDelete
    Replies
    1. The Chinese voices you're referring to may be provided by your Operating System, and not part of Read Aloud. Read Aloud only provides the 1 GoogleTranslate Chinese, and 5 Microsoft Chinese voices. I have just tested and they all appear to work fine. The Google 國語(臺灣) voices are provided by the Chrome browser, they should work as well.

      Delete
  38. This comment has been removed by the author.

    ReplyDelete
  39. Hi there! I've been using Read Aloud for a year or so before coming across the possibility of using Polly or Wavenet voices and as someone with dyslexia it's fantastic to be so much more inclined to use TTS throughout my day. Thanks so much for the work you've put into this project and being so engaged with the users.

    I was aware of Poly and wavenet voices but didn't know about their free tiers and had to think it would be prohibitively expensive to use on a personal basis like this.

    I've signed up for GCP and set things up using the free tier with the $300 promo but as I'm not a developer and haven't used any of these platforms before I was just a bit concerned about running into any unexpected charges because I hadn't set things up correctly.

    I went in and closed the billing account on GCP associated with the wavenet program I created to get the API which I'm hoping is all that I'd need to do if I managed to go beyond the 1 million character limit that month included within the free tier. Perhaps I shouldn't try using your extension to turn the Google Play Books i have into audiobooks read with wavenet voices as I can't imagine I could use more than 1 million characters any other way but do you know if my assumptions here are correct?

    Cheers!

    ReplyDelete
    Replies
    1. Hello, indeed you don't want to overrun that free tier. You can go to https://aws.amazon.com/polly/pricing/ to see approximately how much 1 million characters is. I wouldn't recommend using it to read books. And ideally you'd want to convert the whole book to speech once, and save the audio to listen to it again and again, possibly share with friends as well. I recommend using it for casual reading only, news, etc.

      Delete
  40. Hi, thanks for the nice app. My default US voice changed from male to female today which is okay, but now within a reading it is randomly switching back to male for a sentence and then back to female. Is there any support for this? I tried uninstalling and reinstalling already. Thanks. (Again, the voice I'm looking to use is the default Google US voice).

    ReplyDelete
    Replies
    1. Ah, unfortunately we don't have control over that. Please install this voice: https://chrome.google.com/webstore/detail/us-english-female-text-to/pkidpnnapnfgjhfhkpmjpbckkbaodldb . After installing, it will appear in the voice list as "US English Female TTS (by Google)", and it works offline as well.

      Delete
  41. Hi, I have the Amazon AWS polly 12-month free tier, I would like add the Matthew, Male voice to the extension, but don't know how.

    ReplyDelete
    Replies
    1. Please right click the Read Aloud icon and choose Options. Then in the Voice dropdown, scroll to the bottom and select "Enable Custom Voices". Enter your AWS access_key_id and secret_access_key and click Save. The polly voices will appear as "AmazonPolly Matthew", etc.

      I assume you know how to create the access_key_id and secret_access_key, i.e. you need to goto IAM, create a user with programmatic access, and grant that user the AmazonPollyReadOnly permission.

      Delete
    2. Had to come back and say thanks! Couldn't figure out what I was doing wrong, had to restart Chrome so the changes would take effect. Matthew +neural and +newscaster are amazing voices. I use the extension daily and want to say thank you for helping me with my dyslexia. I want to buy you some coffee, please check your payPal.

      Delete
    3. Yea, the neural and newscaster voices are amazing. Thank you for your financial support!

      Delete
    4. Also, beware that the free tier only covers 1 million neural characters. You might run up the bill pretty quickly after that

      Delete
  42. Great extension. Is there a way I can use my own IBM Watson API with your extension. I would really appreciate if you could guide me about how can i achieve it.

    ReplyDelete
    Replies
    1. Not at the moment. But I don't think you'd want to do that. IBM's free tier only includes 10,000 characters. After that it's really expensive ($20 per million characters), not suitable for our purpose.

      We're aware the IBM voices has stopped working recently, but we expect it to be restored within this week when release 1.30.1 is approved. So please stay tuned.

      Delete
  43. I just saw your new post saying that you had to discontinue IBM Watson voices. I think I can afford it with the rate that you've mentioned above. I would really appreciate if you could let me know where and in which file of your extension I can add my own IBM Watson API URL and API Key to make it work on my local computer? Or if there's any other possible solution you know to use IBM Watson voices? Thanks in advance!

    ReplyDelete
    Replies
    1. Yes, it's quite unfortunate. I'll add support for using your own IBM Cloud account soon. So you can use your own API key. Thanks

      Delete
    2. This comment has been removed by the author.

      Delete
  44. why can,t i get it for free

    ReplyDelete
  45. Excellent. I struggled a little to set up my GCP TTS with wavenet, but finally it worked. Amazing job guys. I think I have an issue, possibly dyslexia, I don't know, but certainly it is much easier for me to read while listening what I am reading. A good quality natural voice is always much more preferable than a robotic one. I was amazed about the GCP voices, Wavenet AI really is amazing!

    ReplyDelete
  46. Awesome app. It works way better than natural reader. Is there a way to get it to autoscroll the document or highlighlight in the acutal document, instead of the popup window? (PC Chrome, Google Docs) Thanks

    ReplyDelete
    Replies
    1. Sorry, it cannot autoscroll/hightlight directly in the document at this time. It's somewhat difficult to do.

      Delete
  47. I purchased the premium voices 2 times meaning i have 18,000,000 available caracters. But when i try and use the Amy voice it says i cant? Why is that and how do i fix this?

    ReplyDelete
    Replies
    1. I think I saw your message on Chrome Web Store. You said "Extensions manifest must request permission to access the respective host". Make sure you are on a http:// or https:// URL before starting Read Aloud. Extensions cannot access content inside other extensions, or in the Reader mode, etc. If you need further help, please email support@lsdsoftware.com

      Delete