Help Center

Open AI

Note: These help docs are a work in progress and are constantly being updated for clarity and completeness.

The settings below are related to the Artificial Intelligence app.

Information
  • The following settings are compatible with OpenAI: Smart reply, Human takeover, Disable for the tickets area, Disable during office hours, Bot response delay, Reply to user emails, Reply to user text messages.
  • If a human takes control, the OpenAI chatbot is deactivated, and is remains deactivated also if the agent goes offline.
  • The default model is gpt-3.5-turbo.
  • OpenAI supports the use of voice messages through speech recognition.

Settings

See information about most OpenAI settings here.

Chatbot mode

This feature is related to the Settings > Artificial Intelligence > OpenAI > Chatbot mode option. It allows you to configure the information OpenAI will use to respond to user messages.

  • Only general questions OpenAI will respond only to general questions and questions related to previous user messages in the same conversation, without using any information you provide. This is the default setting.
  • Only questions related to your sources OpenAI will only respond to questions related to the information you provide. The user messages will be ignored. You have to first train the chatbot with your sources.
  • All questions OpenAI will only to questions related to the information you provide, to general questions, and to questions related to previous user messages in the same conversation. You have to first train the chatbot with your sources. This is the recommended mode.

Chatbot

The OpenAI chatbot feature functions by providing automated responses to user messages. Select the Chatbot mode setting to specify the questions that the chatbot is capable of responding to. Utilize the human takeover feature to enable the chatbot to redirect the chat to a human agent as necessary.

Fallback message

The fallback message is sent when OpenAI is unable to understand the user question.

Prompt

The prompt instructs OpenAI on how to respond by providing relevant information that can be utilized to answer user inquiries. To comprehend the process, refer to the example prompts listed below.

If you choose to activate the human takeover feature, you need to tell OpenAI to respond with I don't know if it is unable to provide an answer to the user's question. You should always include the following text in the prompt: Respond "I don't know", if not sure about the answer.

Prompt - Message rewriting

This prompt instructs OpenAI on how to rewrite a message when the Message rewrite button is active. This prompt should be in English. Generous Chat automatically add the following text when required: and use the user langauge, add greetings.

Replies from the training sources can be utilized with the prompt feature as they are compatible.

Spelling correction

This feature automatically fix any spelling mistakes in the agent's message.

Smart reply

This feature allow to enable the Smart Reply feature also if the chatbot is not active. If the chatbot is active the Smart Reply feature will use OpenAI automatically.

Message rewrite button

This feature adds a button to the text field of the conversation area, click on it to rewrite your message and make it more friendly and professional. Greetings will be added automatically if there is no previous agent messages. If your language is not English, you have to edit your agent/admin profile and set the correct language.

Speech recognition

When this option is active, audio messages will be converted to written text.

  • You do not need to activate this feature for the chatbot to understand audio messages. The speech recognition feature of the chatbot is always enabled.
  • For this feature to work better, the user language should be know. To automatically detect the user's language, you can enable the language detection feature.
  • The text will be displayed alongside the audio player.
  • The generated text is compatible with the translation features.

Source links

When this option is active, the response will automatically include the links to the sources and training data used for generating the answer. To include articles as sources you have first to set the articles page in Settings > Articles > Articles page URL.

Note data scraping

When this option is active, you will see a menu in the notes panel. The menu enables the automatic extraction of specific information from the user's messages.

OpenAI parameters

Generous Chat allows you to adjust various OpenAI parameters such as temperature and logit_bias, for more details visit https://platform.openai.com/docs/api-reference/completions/create. Don't set any values if you don't know what these parameters do or OpenAI may stop working.

Logit bias

The Logit bias parameter must be JSON string and the keys must be strings, for example: {"2435":-100, "640":-100}.

Generate user questions

When this option is active, OpenAI will generate new user questions in real time when you open the chatbot training window.

Training

With this feature, your chatbot can undergo training using your website, texts, PDF documents and more. Once the training is successfully completed, the chatbot will be able to answer questions related to your contents. To access this feature, navigate to Chatbot > Training.

  • To initiate the chatbot training, enter the chatbot area. It is possible to enter URLs for websites, text or PDF files, or XML sitemaps.
  • Once the sources are set, click the Train chatbot button and await completion of the training process.

Information

  • You can add and manage personalized question and answers from Chatbot > Training > Q&A and from the chatbot training window.
  • It is only possible to upload files in PDF and TXT formats.
  • If you retrain the same website, you must first delete the previous website's training data; otherwise, only new URLs will be trained.
  • You can provide the website URL and all child URLs will be included and crawled, but with large websites, it is more efficient and less prone to errors and infinite link loops to utilize an XML sitemap instead of relying on the website URL. You can create it with a service like https://www.xml-sitemaps.com.
  • If you want to train your chatbot using specific pages from your website instead of all of them, you can make use of an XML sitemap. Create one using a tool like https://www.xml-sitemaps.com, and then remove the pages you do not wish to include by editing the file in a text editor. To use the XML sitemap, you need to upload it either onto your server or an external online location. Afterward, add the URL of the sitemap in Chatbot > Training > Website. If your sitemap contains more than 1000 URLs, it's advisable to split it into multiple files and train the chatbot with one file at a time to prevent server crashes.
  • You can upload large files and your XML sitemap with a service like https://tmpfiles.org.
  • As soon as the training is completed, the uploaded files are removed.
  • To add new training sources, simply train the chatbot again. The previous training sources will not be lost, and only the new sources will be added.
  • There are character limits for training the chatbot. You can view the character limits in your membership page as per your selected plan.
  • The embedding model is essential for training your chatbot and handling all user messages. We currently use the text-embedding-3-small model. It is necessary for these scenarios and cannot be disabled or changed. You can find pricing information at https://openai.com/pricing. Check out the pricing for the text-embedding-3-small model in the Embedding modelssection.
  • The responses generated by OpenAI have the feature to include in the reply a link to the corresponding website page where the answer was sourced.
  • Go to Chatbot > Training > Information, and click the Delete all training data button to remove all previous training data for the chatbot.
  • To allow your chatbot to respond to questions related to real-time data, use the Google search feature.

Training via cron job

You can enable automatic training based on timed intervals, just enable the Settings > Artificial Intelligence > OpenAI > Training via cron job option and the cron job will be activated.

Multilingual training

If your user base is multilingual, you can train the chatbot with content in multiple languages and limit the chatbot to retrieve answers only from the sources in the user's language. To activate this feature, check the Settings > Artificial Intelligence > OpenAI > Multilingual Training Sources option.

  • Files Files currently support only one language.
  • Websites The language of the website is detected automatically. For Generous Chat to comprehend the language of your web pages, the <html> must contain the attribute lang.
  • Q&A Q&A currently support only one language.
  • Articles All article languages are used automatically.
  • Conversations If automatic translation is active, only messages in the agent's language will be used.

Q&A

The information below is related to the Question and Answers section of the chatbot training area. Add questions and answers to the chatbot to improve its performance. The chatbot will use this information to respond to user inquiries.

Question

Enter the user messages that will trigger the answer. Add as many question variations as necessary. For example the questions to the answer I'm a chatbot! could be Who are you?, What are you?, Are you a bot?.

Answer

Enter the text that will be used to answer the user question.

Function calling

Function calling allows you to connect the chatbot to external tools and systems. The chatbot will query your server and return the information from your server to the user. More details here. Our support doesn't include assistance with this feature, as it is for advanced users and requires custom code on your server. For help with this, feel free to hire us emailing support@generous.chat

Parameters
  • URL: Enter the URL of the API endpoint that will supply the necessary values for the function.
  • Headers: Enter key-value header parameters, separated by commas. E.g. apikey:123345, json:true.
  • Properties: Enter values that the user must provide to the chatbot. For example a city, a tracking number, etc. The chatbot will ask the user for these values. If you already know all the possible values, you can enter them in the Allowed values field.
Response from your server

Your server must returns a JSON array with the values required by the chatbot. E.g. ["order_status" => "Delivered 2 hours and 25 minutes ago"].

Set data and actions

Set the specified user values when the question is asked. You will see such values in the user details panel. You can use the following merge fields to assign values extracted from the user messages. Include these fields in the answer and they will be replaced with the actual values: {language}. You can also use this feature to perform actions like assigning departments, agents, and tags to the conversation.

Flows

The information below is related to the chatbot flows area. Flows allows you to esely create conversation flows powered by the chatbot. Use them to guide the user toward a specific goal with a series of pre-defined messages.

Flow blocks

  • Start: Use the start block to set when the flows should start. It can be everytime the user start a new conversation, when it sends a specific message, or on page load. If you set conditions, the flow will only start when all conditions are met.
  • Send message: Send a message to the user.
  • Send button list: Send a list of buttons to the user.
  • Send video: Send a video to the user.
  • Get user details: Get the user details and store them in Generous Chat. You will see such values in the user details panel. You can leave the description field empty for default details, but it's required if using custom user fields.
  • Set data: Set the specified user values when the block is executed. You will see such values in the user details panel.
  • Actions: Execute the specified actions when the block is executed.
  • Conditions: Use it to create different branches in the flow. If the conditions are met, the flow will follow the branch set as true, otherwise the one set as false.
  • Rest API: Use it to send data to your server. The body must be a JSON array and the Generous Chat user details are automatically included in it under the sb key. Use the save responsesection to save specified user details with values from your server response. Use the JSON dot notation. Our support doesn't include assistance with this feature, as it is for advanced users and requires custom code on your server. For help with this, feel free to hire us by emailing support@generous.chat.
Rate and review
HelpfulNot helpful