This project combines multiple operations in Microsoft Azure Cognitive Services into one GUI, including QnA Maker, LUIS, Computer Vision, Custom Vision, Face, Form Recognizer, Text To Speech, Speech To Text and Speech Translation.
It's very user-friendly for users to implement any operation mentioned above. Once the execute button is clicked, the program will connect to the Azure cloud by calling the REST API, and then return the response to the GUI to display the result.
It is written in C# and uses Windows Forms for its graphical user interface (GUI).
| IDE | Visual Studio 2019 |
|---|---|
| .NET Core | .NET Core 3.1 |
| Programming Language | C# |
Please click the following figures or links to watch GUI demo videos:
Microsoft Azure Cognitive Services using C#-中文版
1. Overall Structure
2. Language - QnA Maker in Azure
●Construct the Q&A's knowledge base in Azure QnA Maker service, and then train and test the model, and finally publish the API.
3. Language - LUIS in Azure
●Construct the LUIS's Intents, Entities and the corresponding example sentences in Azure LUIS service, and then train and test the model, and finally publish the API.
4. Language - Config Setting
5. Language - QnA Maker
●Users can easily ask questions and get answers.
6. Language - LUIS
●Analyze a sentence, and then get the intent, entities and the corresponding entity typies.
7. Vision - Computer Vision
●Analyze Image: Analyze the image, and then get some information, including gender, age, summary, categories, tags, adult, color scheme, landmarks, image type, etc.
●Detect Object: Detect all possible objects in the image, and then get the object class and the corresponding confidence.
●Read Text (OCR): OCR means Optical Character Recognition. This operation can detect the texts and url links in the image and display them in the right information field.
8. Vision - Custom Vision
●Custom Vision is classified into two categories: Classification and Object Detection. It allows users to train customized models according to their needs.
●Classification:
●Object Detection:
●Note: Probability Threshold must be properly determined by a large amount of data verification.
9. Vision - Face
●Detect all possible faces in the image, and get the information for each face, such as accessories, age, blur, emotion, exposure, facial hair, gender, glasses, hair, head pose, make-up, noise, occlusion, smile, etc.
10. Vision - Form Recognizer
●Recognize the receipt information in the image.
11. Speech - Text To Speech
●Convert text to speech based on different languages. The output voice can select gender and voice name.
12. Speech - Speech To Text
●Convert speech to text based on different languages.
●Its largest advantage is that it does not need to be assigned a specific language, it will detect which language the voice belongs to by itself.
●Thus, I think this is a practical tool that allows us to practice pronunciation.
13. Speech - Speech Translation
●Speech Translation can translate a speech into multiple languages. The output voice can select gender and voice name.
If you have any questions or suggestions about code, project or any other topics, please feel free to contact me and discuss with me. 😄😄😄
