Script

Clap recognition script with file opening

Простой скрипт на Python, который слушает микрофон и реагирует на двойной хлопок. Когда ловит два хлопка подряд, открывает указанный файл, картинку или программу (в коде просто поменяйте путь и всё). Голос в большинстве случаев не реагирует, но скрипт не идеальный, иногда бывают баги. Например, звуки клавиатуры или резкие щелчки могут считаться хлопками.

import sounddevice as sd
import numpy as np
import os
import time
import subprocess
import platform

THRESHOLD = 0.15
CLAP_TIMEOUT = 2.5
MIN_CLAP_INTERVAL = 0.15
SILENCE_THRESHOLD = 0.05
AUTO_CALIBRATE = True

# укажите имя файла тут
# можно фотографию или любой другой exe
# можно также указать полный путь
IMAGE_FILENAME = "dota2.png" #это просто пример(я делал тест с этой фоткой)

MAX_CLAP_DURATION = 0.2
MIN_CLAP_PEAK = 0.25

last_clap_time = 0
clap_count = 0
in_clap = False

background_noise = 0.0
dynamic_threshold = THRESHOLD

sound_start_time = 0
sound_volumes = []

selected_device = None
image_path = None


def desktop():
return os.path.join(os.path.expanduser("~"), "Desktop")


def find_image(filename):
paths = [
desktop(),
os.path.expanduser("~"),
os.path.expanduser("~/Pictures"),
os.path.expanduser("~/Downloads"),
os.getcwd(),
]

for p in paths:
fp = os.path.join(p, filename)
if os.path.exists(fp):
return fp

return None


def select_microphone():
devices = sd.query_devices()
inputs = [i for i, d in enumerate(devices) if d["max_input_channels"] > 0]

if not inputs:
print("Нет доступных микрофонов")
return None

for i in inputs:
print(f"[{i}] {devices[i]['name']}")

choice = input("Микрофон (Enter — по умолчанию): ").strip()

if choice == "":
return sd.default.device[0]

try:
idx = int(choice)
if idx in inputs:
return idx
except ValueError:
pass

print("Неверный выбор")
return sd.default.device[0]


def calibrate(device):
global background_noise, dynamic_threshold

volumes = []

def cb(indata, frames, time_info, status):
volumes.append(np.abs(indata).max())

with sd.InputStream(device=device, callback=cb, channels=1, samplerate=44100):
time.sleep(2)

if not volumes:
return False

background_noise = np.mean(volumes)
dynamic_threshold = max(0.1, background_noise * 3)
return True


def open_file(path):
if not os.path.exists(path):
print("Файл не найден:", path)
return

try:
if platform.system() == "Windows":
os.startfile(path)
elif platform.system() == "Darwin":
subprocess.run(["open", path])
else:
subprocess.run(["xdg-open", path])
except Exception as e:
print("Ошибка открытия файла:", e)


def audio_callback(indata, frames, time_info, status):
global clap_count, last_clap_time, in_clap
global sound_start_time, sound_volumes

volume = np.abs(indata).max()
now = time.time()
threshold = dynamic_threshold

if volume > threshold:
if not in_clap:
in_clap = True
sound_start_time = now
sound_volumes = [volume]
else:
sound_volumes.append(volume)
if now - sound_start_time > MAX_CLAP_DURATION:
in_clap = False
sound_volumes = []

elif volume < SILENCE_THRESHOLD and in_clap:
duration = now - sound_start_time
peak = max(sound_volumes) if sound_volumes else 0

if duration <= MAX_CLAP_DURATION and peak >= max(MIN_CLAP_PEAK, threshold * 1.5):
delta = now - last_clap_time

if delta > MIN_CLAP_INTERVAL:
clap_count = clap_count + 1 if delta < CLAP_TIMEOUT else 1
last_clap_time = now
print(f"Хлопок {clap_count}")

if clap_count == 2:
open_file(image_path)
clap_count = 0
time.sleep(0.3)

in_clap = False
sound_volumes = []


def main():
global selected_device, image_path

image_path = find_image(IMAGE_FILENAME)
if not image_path:
print("Файл не найден")
return

selected_device = select_microphone()
if selected_device is None:
return

calibrate(selected_device)

print("Ожидание двух хлопков")
print("Файл:", image_path)

try:
with sd.InputStream(
device=selected_device,
callback=audio_callback,
channels=1,
samplerate=44100,
blocksize=1024,
dtype=np.float32,
):
while True:
time.sleep(0.1)
except KeyboardInterrupt:
print("Остановлено")


if __name__ == "__main__":
main()

Python

import sounddevice as sd

import numpy as np

import os

import time

import subprocess

import platform



THRESHOLD = 0.15

CLAP_TIMEOUT = 2.5

MIN_CLAP_INTERVAL = 0.15

SILENCE_THRESHOLD = 0.05

AUTO_CALIBRATE = True



# укажите имя файла тут

# можно фотографию или любой другой exe 

# можно также указать полный путь

IMAGE_FILENAME = "dota2.png" #это просто пример(я делал тест с этой фоткой)



MAX_CLAP_DURATION = 0.2

MIN_CLAP_PEAK = 0.25



last_clap_time = 0

clap_count = 0

in_clap = False



background_noise = 0.0

dynamic_threshold = THRESHOLD



sound_start_time = 0

sound_volumes = []



selected_device = None

image_path = None





def desktop():

    return os.path.join(os.path.expanduser("~"), "Desktop")





def find_image(filename):

    paths = [

        desktop(),

        os.path.expanduser("~"),

        os.path.expanduser("~/Pictures"),

        os.path.expanduser("~/Downloads"),

        os.getcwd(),

    ]



    for p in paths:

        fp = os.path.join(p, filename)

        if os.path.exists(fp):

            return fp



    return None





def select_microphone():

    devices = sd.query_devices()

    inputs = [i for i, d in enumerate(devices) if d["max_input_channels"] > 0]



    if not inputs:

        print("Нет доступных микрофонов")

        return None



    for i in inputs:

        print(f"[{i}] {devices[i]['name']}")



    choice = input("Микрофон (Enter — по умолчанию): ").strip()



    if choice == "":

        return sd.default.device[0]



    try:

        idx = int(choice)

        if idx in inputs:

            return idx

    except ValueError:

        pass



    print("Неверный выбор")

    return sd.default.device[0]





def calibrate(device):

    global background_noise, dynamic_threshold



    volumes = []



    def cb(indata, frames, time_info, status):

        volumes.append(np.abs(indata).max())



    with sd.InputStream(device=device, callback=cb, channels=1, samplerate=44100):

        time.sleep(2)



    if not volumes:

        return False



    background_noise = np.mean(volumes)

    dynamic_threshold = max(0.1, background_noise * 3)

    return True





def open_file(path):

    if not os.path.exists(path):

        print("Файл не найден:", path)

        return



    try:

        if platform.system() == "Windows":

            os.startfile(path)

        elif platform.system() == "Darwin":

            subprocess.run(["open", path])

        else:

            subprocess.run(["xdg-open", path])

    except Exception as e:

        print("Ошибка открытия файла:", e)





def audio_callback(indata, frames, time_info, status):

    global clap_count, last_clap_time, in_clap

    global sound_start_time, sound_volumes



    volume = np.abs(indata).max()

    now = time.time()

    threshold = dynamic_threshold



    if volume > threshold:

        if not in_clap:

            in_clap = True

            sound_start_time = now

            sound_volumes = [volume]

        else:

            sound_volumes.append(volume)

            if now - sound_start_time > MAX_CLAP_DURATION:

                in_clap = False

                sound_volumes = []



    elif volume < SILENCE_THRESHOLD and in_clap:

        duration = now - sound_start_time

        peak = max(sound_volumes) if sound_volumes else 0



        if duration <= MAX_CLAP_DURATION and peak >= max(MIN_CLAP_PEAK, threshold * 1.5):

            delta = now - last_clap_time



            if delta > MIN_CLAP_INTERVAL:

                clap_count = clap_count + 1 if delta < CLAP_TIMEOUT else 1

                last_clap_time = now

                print(f"Хлопок {clap_count}")



                if clap_count == 2:

                    open_file(image_path)

                    clap_count = 0

                    time.sleep(0.3)



        in_clap = False

        sound_volumes = []





def main():

    global selected_device, image_path



    image_path = find_image(IMAGE_FILENAME)

    if not image_path:

        print("Файл не найден")

        return



    selected_device = select_microphone()

    if selected_device is None:

        return



    calibrate(selected_device)



    print("Ожидание двух хлопков")

    print("Файл:", image_path)



    try:

        with sd.InputStream(

            device=selected_device,

            callback=audio_callback,

            channels=1,

            samplerate=44100,

            blocksize=1024,

            dtype=np.float32,

        ):

            while True:

                time.sleep(0.1)

    except KeyboardInterrupt:

        print("Остановлено")





if __name__ == "__main__":

    main()

К слову, писал этот код на версии 3.14.2

Зависимости

pip install sounddevice numpy