Addressing Your Questions

1. Is Pandorabot an outdated AI? If so, can you recommend another AI that is compatible with ARC, and provide instructions on how to install and configure it?

Pandorabot is still functional but may not be as advanced as some newer AI options. Given your current setup, I recommend using DialogFlow by Google, which you already have installed. DialogFlow is a powerful AI that can handle natural language processing and is well-integrated with ARC.

Steps to Configure DialogFlow:

1. Create a DialogFlow Agent:

   • Go to the DialogFlow Console.
   • Click on "Create Agent" and fill in the necessary details.

2. Set Up Intents:

   • Create intents that define how your robot should respond to various inputs.
   • For example, create an intent for greetings, commands, etc.

3. Integrate DialogFlow with ARC:

   • In ARC, go to the DialogFlow skill configuration.
   • Add your Google Service Account JSON file and Project ID.
   • Use the ControlCommand to send phrases to DialogFlow and handle responses.

4. Example Script:

   if (getVar("$APIIntent") == "greeting") {
       Audio.say("Hello! How can I assist you today?");
   } else {
       Audio.say(getVar("$APIResponse"));
   }
   

For detailed instructions, refer to the DialogFlow Skill Manual.

2. I receive responses through both the PC speakers and the IoTiny speakers simultaneously. I cannot choose which speakers to use.

To control which speakers are used for audio output, you can specify the output device in your script. Since you are using Python, you can use the Audio.sayEZB() function to direct the audio to the IoTiny speakers.

Example Script:

Audio.sayEZB(getVar("$pandoraResponse"));

If you want to use the PC speakers, you can use:

Audio.say(getVar("$pandoraResponse"));

Additional Resources

• DialogFlow Manual: DialogFlow Skill Manual
• Watson Text To Speech Manual: Watson Text To Speech Manual
• IoTiny Hardware Manual: EZ-B IoTiny Manual

Encouragement

Feel free to explore the full potential of ARC by integrating various robot skills and customizing your scripts. If you have any more questions or need further assistance, don't hesitate to ask!

Speaking

As Athena pointed out, the script command determines the audio output device. Two commands can be referenced in the support section for the specific manual pages. I can't provide a list of every support page, as it is extensive, but here are the relevant ones to get you started.

sayEZB: https://synthiam.com/Support/javascript-api/Audio/sayEZB (that speaks out of the EZB speaker)

say: https://synthiam.com/Support/javascript-api/Audio/say (that speaks out of the PC speaker)

The support section will be helpful if you prefer to look up the commands without waiting for Athena to respond.

AI

In our opinion, the best AI is from our LLM partner, Open AI, with the product Chat GPT. You can use the robot skill from Open AI here: https://synthiam.com/Support/Skills/Artificial-Intelligence/OpenAI-ChatGPT?id=21559

You will find several sources in the news about Open AI leading the AI revolution over the last several years. We find their product superior to the competition. However, if you wish to use other LLMs, the Open AI protocol is generally compatible, and the end point URL can be altered in the configuration settings. The manual for the above-mentioned robot skill explains more.

@synthiam.support   I am playing now with aimlbot but it seems that it uses the same information as Pandora does. I will follow your advice with the robot skill open ai chat gpt. I'd like to check the other skills, camera  cognitive face and vision .  thanks for your support

AIMBot uses the same technology as pandorabot, I believe. It's AIML, which is what pandora bot says their system uses. The AIMLBot runs local vs in the cloud, which is the only difference. I always have a good chat experience with open AI chat gpt robot skill - plus you can assign actions and such to it.