Hey everybody, I’m new here. I joined because I was hoping somebody could help me with a simple chatbot I’m making? (And so I can learn more about making these) I’m a newb to chat bots, just started this week in fact.

It’s just a slightly tweaked bot I have on Pandorabots. Extremely simple. It’s made for Mega Man fan community purposes. I want to somehow give it speech recognition, and text-to-speech.

It already has a bar where it can turn your text into speech, but it doesn’t submit it as the conversation bar. Can somebody give me the proper html coding for this? And html coding for the bot’s text to come out as speech as well?

Here is a link to the bot, It’s named Aria.EXE (Mega Man related name)
http://pandorabots.com/pandora/talk?botid=d42f863fee349668

The picture is a question mark because I’m currently holding a contest for other Mega Man fans to design what she looks like. If you don’t know what Mega Man is, it’s a fantastic video game. I suggest you look it up

> http://www.quora.com/Speech-to-Text/What-open-sourced-and-accurate-speech-to-text-engines-and-APIs-currently-exist

See my recent Quora answer to: “What open-sourced and accurate speech-to-text engines and APIs currently exist?”

> http://www.sitepal.com/packages/

SitePal offers a Pandorabots-compatible TTS avatar for “Artificial Intelligence” at $40 per month.

You can get STT (ASR) using the Google (Chrome) Speechrecognition API, see attached screenshot.

Hi Josh
Planet source code is an excellent resource for coding or even ideas on coding.
http://planetsourcecode.com/vb/scripts/BrowseCategoryOrSearchResults.asp?lngWId=10&txtCriteria=text+to+speech
I found these examples for speech synthesis using .net
Usually the only requirement is that you give credit to the originator(s) of the code.

Vince

Thanks guys. I enabled Google’s voice recognition, but the mic doesn’t appear on my bot’s text box. What am I doing wrong?

Are you using Chrome Josh? The speech API is only available in Chrome.

Just to eliminate any misunderstandings, what Steve is saying is that you’ll only see the Speech Recognition microphone when you’re using the Google Chrome web browser.

Also, that only takes care of speech recognition. It doesn’t provide any TTS (Text To Speech) conversion for the chatbot at all. Thus, it’s only doing half the job, and only then, in one specific web browser.

One sneaky way I found to get free and quick TTS was to use the Google translate API. I pass the bot’s output to it and say translate from English to English then speak the results! Not quite sure if it’s in the terms of use but it worked for me.

Chrome is the only browser I use. It’s popping up when I’m on Google.com, but not on my bot.

As for TTS, I think I might have a few answers.

> http://chrome.google.com/webstore/detail/voice-in/gjkogfbjkfchelfjonefnnenhfgglpnn

Try installing this Chrome plugin “Voice In”.

Steve,
Hands down that gets my vote for the “down n dirty” hack of the year award
Vince

Vincent Gilbert - Oct 16, 2012:

Steve,
Hands down that gets my vote for the “down n dirty” hack of the year award
Vince

Yeah, that smacks of “cheating” to me, too. But I was taught early on that cheating is nothing more than using an unexpected resource.

Vincent Gilbert - Oct 16, 2012:

Steve,
Hands down that gets my vote for the “down n dirty” hack of the year award
Vince

Sorry I should have been clearer. Dont get me wrong thats a compliment! Being “Old school” I still use the term “Hack” the way it originated. “A highly creative real world solution to an immediate or pressing problem” LOL I think that this type of cross resource useage is exactly what is intended with making these API(s) available.
The problem with the Internet has always been that you are limited by the controls that are inherent in HTML 4. You can always create an Active X control, or wait for HTML 5. If you use Active X your not cross browser compatible, and if you wait for HTML 5 youll die of old age.

I intend to use Steves “hack” (see above definition for RICH’s online interface
Kudos Steve!

Hehe! I knew what you meant Vincent. After looking around for a TTS solution, this seemed the easiest (and cheapest) method.