Introduction
hello everyone I am here today to introduce my new extension called ChatGPT,
The ChatGPT extension is designed to engage in conversations with the ChatGPT and deliver the resulting response in an API-style structure. It allows MIT App Inventor users to integrate OpenAI's powerful language model into their Android apps seamlessly.
features
Continuous Chat Conversation:
- The code supports continuous chat conversations by sending a list of prompts, enabling dynamic and interactive conversations with the OpenAI bot. so the model don't forget the previous messages that you have sent before
Streaming Support:
- it enables streaming of responses in chunks, making it suitable for handling large responses.
Error Handling:
- The extension includes error handling through the
Error
Block and related error events, which allow you to handle errors gracefully and provide feedback to users.
Audio Transcription:
- The extension includes blocks for audio transcription using OpenAI's Audio Transcriptions, with events for handling and displaying transcribed text.
Audio Translation:
- It also provides functions for audio translation using OpenAI's Audio Translations , with events for processing and displaying translated text.
Blocks
The SendMessage
block is responsible for sending a conversation to the ChatGPT and processing the response. Here's a breakdown of the code:
- Block Description: This Block allows users to interact with the OpenAI ChatGPT and receive structured API-style responses.
- Function Parameters:
prompts
: A list of conversation prompts provided by the user like in the block above .model
: The name of the OpenAI model to be used.apiKey
: The API key for authorization.maxTokens
: The maximum number of tokens in the response.temperature
: A value controlling the randomness of the response.
The RespondedToChat
Block is an event that is triggered when OpenAI provides a response to a user's inquiry within the Chat block. This block carries various parameters, and here is an explanation of each parameter:
responseId
: A string representing the unique identifier for the response generated by OpenAI.responseType
: A string indicating the type of the response object. It typically denotes the data structure used to encapsulate the response.createdTimestamp
: A long value representing the timestamp when the response was created. It is usually in Unix timestamp format.responseModel
: A string indicating the specific OpenAI model that was used to generate the response.choiceIndex
: An integer representing the index of the choice within the response. OpenAI often provides multiple choices, and this parameter indicates which choice is selected.role
: A string indicating the role of the message within the conversation. It can be used to distinguish between different roles, such as "system," "user," or "assistant."content
: A string representing the content of the response. This is the actual text generated by the OpenAI model.finishReason
: A string indicating the reason for the completion of the response. It provides information about why the conversation ended.promptTokens
: An integer representing the number of tokens used in the conversation prompt. Tokens are units of text used by OpenAI models.completionTokens
: An integer representing the number of tokens used in the generated response completion.totalTokens
: An integer representing the total number of tokens used in the entire response, including both the prompt and the completion.
The SendStreamedMessage
function is designed to retrieve a response in chunks from the ChatGPT model. It allows for ongoing communication with the model and is specifically used for streaming responses,
Parameters:
- The function takes several parameters:
id
(integer): An identifier for the stream.prompts
(YailList): A list of prompts (messages) that constitute the conversation with the model.model
(String): The model code used for the conversation.apiKey
(String): The API key required for authentication.maxTokens
(integer): The maximum number of tokens for the response.temperature
(double): A value that controls the randomness of the response.
The StopStream
Block and the associated StoppedStream
event are used in the context of managing streaming operations in the code.
The StoppedStream
Block is an essential component in managing streaming operations and is triggered when the streaming process is manually stopped by calling the StopStream
Block
The GotStream
Block is used to notify when OpenAI has provided a response to a stream request during an ongoing streaming conversation.
The FinishedStream
event is used to notify when all chunks of a stream have been returned through the GotStream
event, indicating the completion of the streaming conversation.
RequestModeration
Description: This function asynchronously requests content moderation using the OpenAI Moderation API. It takes an API key and input text as parameters, sends a POST request to the API endpoint, and processes the response.
Parameters:
apiKey
(String): The API key for accessing the OpenAI Moderation API.input
(String): The input text or content to be moderated.
ModerationResult
Description: This event is triggered when the moderation result is received from the OpenAI Moderation API. It provides information about whether the content is flagged, categories, and category scores as parameters.
Parameters:
flagged
(boolean): Indicates whether the content is flagged.categories
(String): JSON representation of the detected categories.categoryScores
(String): JSON representation of the scores for each category.
Usage: Handle this event to perform actions based on the moderation result, such as updating the user interface or taking appropriate actions based on the moderation outcome.
RequestAudioSpeech
Function
Description: This function is responsible for asynchronously requesting audio speech synthesis from OpenAI's Audio Speech API. It takes various parameters such as API key, input text, model, voice, folder path, and file name. The resulting MP3 content is then written to a file.
Parameters:
-
apiKey
(String): The API key for accessing OpenAI's Audio Speech API. -
text
(String): The input text to be synthesized into speech. -
model
(String): The model to be used for speech synthesis. models tts-1 or tts-1-hd -
voice
(String): The voice to use when generating the audio. Supported voices arealloy
,echo
,fable
,onyx
,nova
, andshimmer
-
folderPath
(String): The path to the folder where the MP3 file will be saved. -
fileName
(String): The name of the MP3 file to be saved.
Exambles -
Alloy :
- Echo :
- Fable :
- Onyx :
You can try other voices
SpeechFileSaved
Event
Description: This event is fired when the MP3 file has been successfully saved. It provides the file path as a parameter.
Parameters:
filePath
(String): The path where the MP3 file has been saved.
Usage: Handle this event to perform actions after the MP3 file has been successfully saved.
SpeechSynthesisError
Event
Description: This event is fired when an error occurs during the audio speech synthesis process. It provides an error message as a parameter.
Parameters:
errorMessage
(String): The error message describing the issue encountered.
Usage: Handle this event to capture and handle errors during the speech synthesis process.
RequestAudioTranscription
Block is responsible for making a request to OpenAI's Audio Transcriptions API to transcribe audio from a provided audio file. (Transcribes audio into the input language.)
The Blcok takes four parameters:
-
apiKey
(API key for authentication), -
audioFilePath
(path to the audio file to be transcribed), -
model
(model configuration), you can set it aswhisper-1
-
responseFormat
( The format of the transcript output, in one of these options:json
,text
,srt
,verbose_json
, orvtt
).
The AudioTranscriptionReceived
block, its purpose is to notify the application when audio transcription data is received
The RequestAudioTranslation
block is designed to request audio translation from OpenAI's Audio Translations API and return the "text" value from the response.
Parameters:
The function takes three parameters:
-
apiKey
(API key for authorization), -
audioFilePath
(path to the audio file to be translated), -
model
(the model used for translation).
This ReturnAudioTranslation
event is triggered when the audio translation response is received.
RequestDALL_EImages
Description
This function initiates a request to the OpenAI DALL-E Images API to generate images based on a given prompt.
Parameters
-
apiKey
(String): The API key for authentication. -
model
(String): (Optional) The model to use for image generation, defaults to "dall-e-2". -
prompt
(String): A text description of the desired image(s) (Required). The maximum length is 1000 characters for dall-e-2 and 4000 characters for dall-e-3. -
n
(int): (Optional) The number of images to generate, defaults to 1. Must be between 1 and 10. For dall-e-3, only n=1 is supported. -
size
(String): (Optional) The size of the generated images, defaults to "1024x1024
". Must be one of "256x256
", "512x512
", or "1024x1024
" for dall-e-2. Must be one of "1024x1024
", "1792x1024
", or "1024x1792
" for dall-e-3 models.
Events
DALL_EImagesGenerated
(List imageUrls): Fired when the DALL-E Images API successfully generates images. Returns a list of image URLs.
DALL_EImagesError
(String errorMessage): Fired when an error occurs during the DALL-E Images API request. Returns an error message.
Function: RequestChatGPTVision(String apiKey, String imageUrl, String prompt)
Purpose: This function sends a request to OpenAI's ChatGPT vision API to analyze an image and provide insights based on the given prompt.
Parameters:
-
apiKey: Your OpenAI API key.
-
imageUrl: The URL of the image to analyze.
-
prompt: A text prompt to guide the analysis (e.g., "What's in this image?").
Function: RequestChatGPTVisionMultipleImages(String apiKey, YailList imageUrls, String prompt)
Purpose: This function sends a request to OpenAI's ChatGPT vision API to analyze multiple images and provide insights based on the given prompt.
Parameters:
-
apiKey: Your OpenAI API key.
-
imageUrls: A YailList containing the URLs of the images to analyze.
-
prompt: A text prompt to guide the analysis (e.g., "Compare these images").
RequestChatGPTVisionFromFile
Purpose: Analyzes a single local image file and provides insights based on a text prompt.
Parameters:
-
apiKey: Your OpenAI API key.
-
imagePath: The file path of the image to analyze.
-
prompt: A text prompt to guide the analysis (e.g., "What's in this image?").
-
detail: The desired level of detail for the analysis (
low
,high
, orauto
). -
maxTokens: The maximum number of tokens allowed in the API response.
RequestChatGPTVisionMultipleImagesFromFile
Purpose: Analyzes multiple local image files and provides insights based on a text prompt.
Parameters:
-
apiKey: Your OpenAI API key.
-
imagePaths: A YailList containing the file paths of the images to analyze.
-
prompt: A text prompt to guide the analysis (e.g., "Compare these images").
-
detail: The desired level of detail for the analysis (e.g., "high").
-
maxTokens: The maximum number of tokens allowed in the API response.
Events:
-
ChatGPTVisionResponseReceived : This event is fired when the API response is successfully received and parsed. It provides the following parameters:
-
id: The unique ID of the response.
-
object: The type of object returned ("chat.completion").
-
model: The model used to generate the response.
-
role: The role of the response ("assistant").
-
content: The main content of the response, containing the analysis of the image.
-
-
ChatGPTVisionError(String errorMessage) : This event is fired if an error occurs during the API request. It provides the error message.
-
The response content will vary depending on the image and the prompt provided.
ChatGPT Extension- Embeddings Functionality
1. GetEmbeddings(String apiKey, String text, String model)
-
Description: This function sends a request to OpenAI's Embeddings API to get the numerical representation (embedding) of a given text.
-
Parameters:
-
apiKey: Your OpenAI API key (required for authentication).
-
text: The text string you want to embed.
-
model: The specific embedding model you want to use (e.g., "text-embedding-ada-002, text-embedding-3-small ,text-embedding-3-lar").
-
-
Functionality:
-
It constructs an API request with your text and the chosen model.
-
It sends this request to OpenAI's server.
-
It then calls the processEmbeddingsAPIResponse function to handle the server's response.
-
-
Events Triggered:
-
EmbeddingsReceived: Fired upon a successful response, containing the embeddings.
-
EmbeddingsError: Fired if an error occurs during the request.
-
2. EmbeddingsReceived(String embeddings)
-
Description: This event is fired when the GetEmbeddings function successfully receives a response from the OpenAI API.
-
Parameter:
- embeddings: The text's embedding, returned as a string representation of a list of numbers.
3. EmbeddingsError(String errorMessage)
-
Description: This event is fired when an error occurs at any point during the embedding request process.
-
Parameter:
- errorMessage: A descriptive error message to help you understand the issue.
preview :
I also use this Extension in this project :
Aix file :
you can purchase the AIX and the AIA file from here via PayPal the two files cost 2$ for limited period of time after you pay you will be automatically redirected to the download URL of the extension