Видео

Please reply soon

We are using a pre built model that is already trained.

pyobjc is not for windows it only works with macos i have the same problem wondering if u found the soulution

I’m getting reports it’s working on M3. Double check installations and source code: github.com/Kent-Taylor/object-detection/blob/main/main.py

pip install tensorflow

Wow you were so close! ❤️

So saying “okay Google” works on Google phones because their code IS always running. Saying “hey Siri” works on iPhones because their code is always running. For those who do have iPhones, when your battery is low you can go into “Low Power Mode”, then saying “hey Siri” doesn’t work anymore, you have to manually hold down the home or power button. That’s because they turn off the “hey siri” program to save power.

@@SalteeKiller okay sir, i mean here i am making a voice controlled RC car project, and i want to implement your code. i need a voice activation before giving commands to the robot like turn right or left, what should i add in your code so that the code runs continuously and gets voice activation

Yeah sure, of course I can’t do it all for you for all your specifications but To implement continuous voice activation before giving commands to the robot, you can modify the `get_audio()` function to continuously listen for the activation phrase before accepting other commands. Here's how you can do it: ```python def get_audio(): while True: recorder = sr.Recognizer() with sr.Microphone() as source: audio = recorder.listen(source) try: speech_text = recorder.recognize_google(audio) print(speech_text) if "R2-D2" in speech_text: speech("\tHow may I help you?") playsound("./sounds/activate.wav") # Now listen for the actual command after activation while True: audio = recorder.listen(source) command = recorder.recognize_google(audio) if command.strip() != "": return command.strip() else: break except sr.UnknownValueError: print("Google Speech Recognition could not understand audio") except sr.RequestError as e: print("Could not request results from Google Speech Recognition service; {0}".format(e)) ``` This modified function first listens for the activation phrase ("R2-D2"). Once it detects the activation phrase, it prompts the user with a response and then listens for the actual command continuously until one is detected. Once a command is detected, it returns that command for further processing in your code.

@@SalteeKiller wow sir, thank you so much

The `cv.detect_common_objects()` function from `cvlib` is designed to work on both Windows and macOS. The error you're encountering might be due to a different issue, such as an incorrect installation or an issue with the camera feed. To troubleshoot the error, you can try the following: 1. **Check Installation:** Ensure that `cvlib` is installed correctly in your Python environment. You can reinstall it using pip: ``` pip install cvlib ``` 2. **Check Camera Feed:** Make sure that your camera is connected properly and that it's providing frames to the script. You can print out the `frame` variable before calling `cv.detect_common_objects()` to check if it's `None`. 3. **Update Code:** Try updating your code to handle potential errors more gracefully. For example, you can add error handling around the `cv.detect_common_objects()` call to catch any exceptions: ```python try: bbox, label, conf = cv.detect_common_objects(frame) except Exception as e: print("Error detecting objects:", e) continue # Skip to the next iteration of the loop ``` This will print out any errors that occur during object detection and continue to the next iteration of the loop without crashing the script.

Sure thing! The error `AttributeError: 'NoneType' object has no attribute 'shape'` is likely occurring because the `frame` variable is `None` at some point in your code. This might happen when the video feed is not being read correctly or when the object detection function (`cv.detect_common_objects`) fails to detect any objects. Here are some steps to handle this issue: 1. **Check Video Capture:** Make sure the video feed is being read correctly from your camera (`VideoCapture(1)`). You can add a check to ensure that the `frame` is not `None` before proceeding with object detection. 2. **Handle NoneType:** Update your code to handle the case where `frame` might be `None`. This can happen if the camera is not providing frames properly or if there's an issue with the connection. Here's an updated version of the code with error handling for `NoneType`: ```python import cv2 import cvlib as cv from cvlib.object_detection import draw_bbox from gtts import gTTS from playsound import playsound from food_facts import food_facts def speech(text): print(text) language = "en" output = gTTS(text=text, lang=language, slow=False) output.save("./sounds/output.mp3") playsound("./sounds/output.mp3") video = cv2.VideoCapture(1) labels = [] while True: ret, frame = video.read() if frame is not None: # Bounding box. # the cvlib library has learned some basic objects using object learning # usually it takes around 800 images for it to learn what a phone is. bbox, label, conf = cv.detect_common_objects(frame) output_image = draw_bbox(frame, bbox, label, conf) cv2.imshow("Detection", output_image) for item in label: if item in labels: pass else: labels.append(item) if cv2.waitKey(1) & 0xFF == ord("q"): break # If no objects are detected, handle the case gracefully if len(labels) == 0: print("No objects detected.") else: i = 0 new_sentence = [] for label in labels: if i == 0: new_sentence.append(f"I found a {label}, and, ") else: new_sentence.append(f"a {label},") i += 1 speech(" ".join(new_sentence)) speech("Here are the food facts I found for these items:") for label in labels: try: print(f" \t{label.title()}") food_facts(label) except: print("No food facts for this item") # Release the video capture and close any open windows video.release() cv2.destroyAllWindows() ``` This code will ensure that if `frame` is `None`, it will skip the object detection step and print "No objects detected." This should help prevent the `NoneType` error from occurring.

If you're encountering an error stating "no module named tensorflow" even though you're not directly using TensorFlow in your code, it might be due to `cvlib` or other libraries indirectly requiring TensorFlow as a dependency. When you install `cvlib`, it installs `tensorflow` as one of its dependencies. So, even if you're not explicitly importing TensorFlow in your code, `cvlib` utilizes it internally for certain operations. To resolve this error, you'll need to ensure that `tensorflow` is installed in your Python environment. You can install it using pip: ``` pip install tensorflow Or if in pip3 then use: pip3 install tensorflow ``` Make sure to install TensorFlow in the same Python environment where you're running your script. After installing TensorFlow, the error should be resolved, and you should be able to run your code without encountering the "no module named tensorflow" error.

@@SalteeKiller cv2.error: OpenCV(4.9.0) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\darknet\darknet_io.cpp:705: error: (-215:Assertion failed) separator_index < line.size() in function 'cv::dnn::darknet::ReadDarknetFromCfgStream' can you define this eror it comes while running the code

That sounds frustrating! If you’re on pc and running into that issue, it sounds like it doesn’t like playsound. So try accessing your local environments dependencies to run the sound instead:) here’s a simple way of playing an mp3 file like we did: import os def play_sound(filename): os.system(f'start {filename}') play_sound('sound.mp3') # Replace 'sound.mp3' with your sound file

This is PyCharm but can also be used in vsCode

Yes so it usually means it can’t see where you have installed gtts as well as play sound. So double check those. Make sure you do all the correct spellings, including upper casing and such as I did in the video:) I don’t think gtts nor play sound has made any huge updates to cause this to not work anymore as it’s still working for others and myself at this point. So go back and double check WHERE, HOW and with which case sensitivity is needed to install it. Let us know!

@@SalteeKiller will this tutorial work on pycharm?

@@moreflozz yes I did this in PyCharm in the video

@@SalteeKiller okay, thanks!

This is PyCharm

same here did you find a fix?

same here did you get a fix to this problem?

pip install cvlib

Hello sorry about your error, however, it doesn’t look like it’s the entire error message? Usually the cause of the error tells you what’s wrong at the bottom of the error, at least it does in Python.

Great, thinking! After the program runs your command, whatever it is, just add a break, and it will go back to the program listening for R2-D2. After an error, or command runs, have it run the “listen for comman” loop.

@@SalteeKiller thank youuu

Hey @salteekiller What would be the difference between an API and dependencies? Is there any API used here?

It can be classes, functions or api’s. Dependencies are typically modules or libraries that provide additional functionality. They are often accessed through classes or APIs (Application Programming Interfaces) that define how to interact with them. APIs can be part of a library or module, and classes are a way to structure and organize code within those libraries.

@@SalteeKiller thank youu❤️

Could you provide more details about the error message you're encountering while trying to install PyWhatKit? 1: what’s Python version do you have? 2: what command are you typing in to install? 3: what directory are you in while trying to type in that command? 4: what does the error message say exactly?

1.) python version 3.9 2.) Pip Install pywhatkit 3.) I am using pycharm...I am making an voice assistant. 4.)Error : Failed Building Wheel for Pillow. (While I am downloading pywhatkit in terminal it shows building wheel for pillow was failing) then I also downloaded an pillow but no use.. What to do sir..please tell me 🙏🏻..I am facing this from one week

Please tell me what to do...to clear this error please sir..

It seems like the error is related to the Pillow library, a dependency of PyWhatKit. Try installing Pillow separately first using the following command: pip install pillow Make sure you are using an updated version of pip. Once Pillow is installed successfully, attempt to install PyWhatKit again: pip install pywhatkit If the problem persists, you might consider creating a virtual environment and then installing the libraries. Sometimes conflicts between existing packages can cause installation issues. Also, ensure you’re running the commands in an administrator or superuser mode, depending on your system, to avoid permission issues during installation.

1. Image Loading Issue: Ensure that the image is loaded correctly using OpenCV (cv2.imread()). Check the path to the image file and make sure it exists. Verify that the image is loaded and not None. 2. Object Detection Result: If the object detection function doesn’t find any objects in the image, it might return None. Ensure that the object detection method is properly applied to the loaded image and that it’s capable of detecting objects in the given context. Here’s an example of how to handle these cases in your code: import cv2 import cvlib as cv image = cv2.imread('test1.jpg') if image is not None: result = cv.detect_common_objects(image) if result is not None and 'objects' in result: for obj in result['objects']: print(obj) else: print("No objects detected in the image.") else: print("Error loading the image.") This code checks if the image is loaded correctly and then proceeds with object detection. If the image loading fails or if no objects are detected, it provides an appropriate message.

The lag you're experiencing might be due to the processing load when performing object detection in real-time. This happens because object detection is computationally intensive, especially when running it continuously on a live video feed. Here are some strategies to potentially reduce the lag: 1. **Optimize Object Detection:** Experiment with different object detection models or libraries. Some models are lighter and faster but might sacrifice accuracy. For example, YOLOv3 might be faster than YOLOv4 but slightly less accurate. 2. **Frame Skipping:** Process every nth frame instead of every frame. For instance, you might detect objects every 3 or 5 frames instead of every frame to reduce processing load. 3. **Reduce Image Size:** Decrease the resolution of frames if possible. Smaller images require less processing power for object detection. 4. **Hardware Acceleration:** Utilize GPU (if available) for acceleration. Many object detection libraries have GPU support for faster computation. Ensure your code is configured to use the GPU if it's available. 5. **Multithreading/Asynchronous Processing:** Use multithreading or asynchronous processing to separate the camera feed acquisition and object detection. This way, while detection occurs, the camera can continue capturing frames without waiting for the detection process to complete for each frame. 6. **Hardware Upgrade:** If possible, consider upgrading your hardware to a more powerful CPU or utilizing a dedicated machine for this task. Implementing these strategies could help mitigate the lag and improve the consistency of your real-time object detection application. Experimenting with these optimizations should allow you to find the right balance between speed and accuracy for your specific use case.

thank you for the advice

Counting individual fish in a live camera stream can be quite challenging, especially distinguishing between a large number of fish with varying shapes, colors, and movements. While object detection can help identify and localize objects (in this case, fish), counting them accurately might require more specialized techniques. Here’s an approach you might consider: 1. Object Detection: Use the object detection techniques you’ve been working with to identify fish in the video stream. This would involve the use of YOLO, SSD, or similar models to detect and label fish. 2. Tracking: Implement a tracking algorithm to keep track of individual fish as they move across frames. Various tracking algorithms, like centroid tracking or Kalman filtering, can help maintain identities of different objects (fish) across frames. 3. Counting: Develop a counting mechanism that keeps track of unique identities assigned to each fish. As new fish are detected, assign them unique IDs and update their positions through the tracking mechanism. Then, count the unique IDs to determine the total number of distinct fish detected. This process requires combining object detection with object tracking and maintaining a system to differentiate between individual fish. It might involve some trial and error to fine-tune the detection and tracking algorithms for accurate counting, especially in complex environments like a pond with multiple fish. Additionally, the success of this approach might depend on factors such as lighting conditions, water clarity, fish size, and the camera’s positioning. Calibration and adjustments may be needed to account for these variables.

@@SalteeKiller Thank you so very much for putting me on the right path. I'm quite a newbe to python and pycharm, but I will give it my best shot. At least I'm very inspired by your works and reply!!!

Ain't no way you used chatGPT to reply to his comment ​@SalteeKiller