ejbeaty / chattykathy Goto Github PK

Failed to execute 'setItem' on 'Storage': Setting the value of 'chattyKathyDictionary' exceeded the quota.
at saveSpeechToLocalCache

Online: http://web.mtools.digital/blog/211

Thanks for help!

Need a way to pause and resume playing speech.

Thanks!

How do I make this work with AWS Sumerian ? I want a speech sentence with a variable that uses ctx.worldData.obj . I get invalid operation error .

After a server error from AWS, all future attempts at calling ChattyKathy fail until a refresh.
Need error catching to deal with this.
Will need to work on it on our end... suggestions welcome...
Thanks

Thanks for the fantastic repo.

I have code that will correctly output speech audio when a button is pressed using kathy.speak() as in the repo example, using desktop chrome browser. On iOS (chrome and safari), no audio is played when the button is pressed, and the following console error is displayed:

Unhandled Promise Rejection: NotAllowedError: The request is not allowed by the user agent or the platform in the current context, possibly because the user denied permission.

This appears to be because the user action of hitting a button is not correctly link to the playing of the sound. This is described, eg here:
scottschiller/SoundManager2#178

Is there a solution to this?

Thanks!

I was looking for something like this... Originally my code was on PHP... so this means I can now store it on a CDN... GREAT!!!

I made a couple of files based on your code, just by changing the VOICES to a female one, and the other file with a male voice.

Now, from your perspective... is there a way I could create conversations by using two voice at diffent given times... Like a dialogue or narration by two people?

Can this be done with your current setup?

Thanks

PS.- How many words can each voice handle per submission?

Support for Safari and mobile browsers requested.

On android is not working.

I get following error:
DOMException: play() can only be initiated by a user gesture. (anonymous) @ ChattyKathy.js:177 playAudio @ ChattyKathy.js:166

When using multiple objects for different voices, if you do Amy.Speak('blah') and then do something like Justin.Speak('blah), Justin uses Amy's voice too.

First: thank you for this outstanding wrapper!

Currently I use the following code:
var awsCredentials = new AWS.Credentials('secret', 'secret2');

This is hardcoded, and it links to a User that has ReadOnly access to Polly only, so even if this is "hacked", the damage won't be too big.

That said, I want to do this right, but the code on the front page is apparently not the whole story.

What should I add, exactly, instead of the above to have all users on my website become the above user, so that this account is basically used as one big guest account? I tried to find out by browsing the AWS help files, but got totally lost.

Dear,
I can save the mp3 audio files, for each message fragment of a long text.
I want to know if Amazon poly can provide any kind of identifier for each mp3 that it creates? And how it can be related to specific text/message?
Or we have to implement it ourselves?

Thanks in advance,

SSML not working: http://docs.aws.amazon.com/polly/latest/dg/ssml-to-speech-console.html

kathy.Speak('<speak>He was caught up in the game.<break time="1s"/> In the middle of the 10/3/2014 <sub alias="World Wide Web Consortium">W3C</sub> meeting he shouted, "Score!" quite loudly. When his boss stared at him, he repeated <amazon:effect name="whispered">"Score"</amazon:effect> in a whisper.</speak>');

On line 111 the TextType has to be set:
``
Text: message,

TextType: 'ssml',

VoiceId: settings.pollyVoiceId
``

I will add a PR with a regex to check if message is ssml and set TextType to ssml.

I'm using Chattykathy and it works great on all browsers, except Safari.
For some reason, Safari doesn't want to play the audio. It gets stuck on the "loadstart"-event and never reaches the "durationchange"-event.

It would be very helpful to have a function that will use this library to save the audio to a local file if running on a server (eg to a local MP3 file) rather than play it through the client browser.
Thanks!