Working bytes stream

README.md

@@ -1,7 +1,3 @@
 # learn_audio
 
 A repo dedicated to audio experiments
-
-# Requirements
-
-One the Ubuntu raspberrypi 20.04 server I needed to have `sudo apt-get install libasound2-dev` installed before run `pip install -r requirements.txt`.

dev_requirements.in

@@ -1,2 +0,0 @@
-black
-pip-tools

dev_requirements.txt

@@ -1,41 +0,0 @@
-#
-# This file is autogenerated by pip-compile with python 3.8
-# To update, run:
-#
-#    pip-compile dev_requirements.in
-#
-black==22.10.0
-    # via -r dev_requirements.in
-build==0.9.0
-    # via pip-tools
-click==8.1.3
-    # via
-    #   black
-    #   pip-tools
-mypy-extensions==0.4.3
-    # via black
-packaging==21.3
-    # via build
-pathspec==0.10.2
-    # via black
-pep517==0.13.0
-    # via build
-pip-tools==6.10.0
-    # via -r dev_requirements.in
-platformdirs==2.5.4
-    # via black
-pyparsing==3.0.9
-    # via packaging
-tomli==2.0.1
-    # via
-    #   black
-    #   build
-    #   pep517
-typing-extensions==4.4.0
-    # via black
-wheel==0.38.4
-    # via pip-tools
-
-# The following packages are considered to be unsafe in a requirements file:
-# pip
-# setuptools

play_audio/simple_play.py

@@ -1,6 +1,8 @@
 import simpleaudio as sa
+import sys
 
-filename = './data/hello.wav'
+filename = sys.argv[1]
+'
 wave_obj = sa.WaveObject.from_wave_file(filename)
 play_obj = wave_obj.play()
 play_obj.wait_done()  # Wait until sound has finished playing

say

@@ -1 +0,0 @@
-

socket_send_recieve/sender.py

@@ -12,7 +12,7 @@ wave_file_path = sys.argv[1]
 if not wave_file_path.endswith('.wav'):
     raise ValueError("File must have .wav file extension.")
 
-(HOST,PORT)=('localhost',9000)
+(HOST,PORT)=('0.0.0.0',9000)
 s=socket.socket(socket.AF_INET,socket.SOCK_STREAM); s.connect((HOST,PORT))
 
 with open(wave_file_path, 'rb') as wave_file:

socket_send_recieve/socket_receiver.py

@@ -3,7 +3,7 @@ import time
 
 import simpleaudio as sa
 
-HOST = "127.0.0.1"
+HOST = "0.0.0.0"
 PORT = 9000
 
 def save_file(client_socket):
@@ -18,9 +18,11 @@ def save_file(client_socket):
 def play_file(conn):
     # you could consider sa.play_buffer ... I don't know if it would be more performant.
     try:
-        wave_object = sa.WaveObject.from_wave_file(conn.makefile(mode="rb"))
+        play_obj = sa.play_buffer(conn.makefile(mode="rb").read(), 1,2,22050)
-        play_obj = wave_object.play()
         play_obj.wait_done()
+        #wave_object = sa.WaveObject.from_wave_file(conn.makefile(mode="rb"))
+        #play_obj = wave_object.play()
+        #play_obj.wait_done()
     except Exception as e:
         print(f"Hit an error: {e} skipping playback")
     return

speech_recog/requirement.in

@@ -1,3 +0,0 @@
-pyaudio
-SpeechRecognition
-gTTs

speech_recog/requirement.txt

@@ -1,26 +0,0 @@
-#
-# This file is autogenerated by pip-compile with python 3.8
-# To update, run:
-#
-#    pip-compile requirement.in
-#
-certifi==2022.9.24
-    # via requests
-charset-normalizer==2.1.1
-    # via requests
-click==8.1.3
-    # via gtts
-gtts==2.2.4
-    # via -r requirement.in
-idna==3.4
-    # via requests
-pyaudio==0.2.12
-    # via -r requirement.in
-requests==2.28.1
-    # via gtts
-six==1.16.0
-    # via gtts
-speechrecognition==3.8.1
-    # via -r requirement.in
-urllib3==1.26.12
-    # via requests

speech_recog/virtual_assistant.py

@@ -1,102 +0,0 @@
-import webbrowser
-import typing
-import os
-import sys
-from abc import ABC, abstractmethod
-
-import speech_recognition as sr
-
-class Action(ABC):
-
-    action_key: str
-
-    @abstractmethod
-    def action_callback(self, action_text) -> typing.Any:
-        pass
-
-    def register_va(self, virtual_assisant):
-        self._va = virtual_assisant
-
-    def sans_action_key(self, action_text):
-        return action_text[len(self.action_key)+1:]
-
-class BrowserAction(Action):
-
-    action_key = "open browser"
-
-    def action_callback(self, action_text):
-        domain = self.sans_action_key(action_text)
-        url = f"https://{domain}"
-        print(f"Opening: {url}")
-        self._va.("Opening {domain}")
-        webbrowser.open(url)
-
-class EchoAction(Action):
-
-    action_key = "echo"
-
-    def action_callback(self, action_text) -> typing.Any:
-        print(action_text)
-        self._va.text_to_speech(self.sans_action_key(action_text))
-
-class StopAction(Action):
-    
-    action_key = "stop listening"
-
-    def action_callback(self, action_text) -> typing.Any:
-        print("Stopping Virtual Assistant...")
-        sys.exit(0)
-
-
-class VirtualAssistant:
-
-    def __init__(self, timeout: int = 3, actions: typing.List[Action]=[]):
-        self.listener = sr.Recognizer()
-        self.timeout = timeout
-        self.action_register = []
-        for action in actions:
-            self.register_action(action)
-
-    def register_action(self, action: Action):
-        """Adds new Action instances to the Virtual Assistant's action register."""
-        self.action_register.append(action)
-        action.register_va(self)
-
-    def parse_action(self, text: str):
-        """Checks the VA command register and executes callbacks when found."""
-        for action in self.action_register:
-            if text.lower().startswith(action.action_key):
-                action.action_callback(text)
-                return
-        print(f"No actions taken on: {text}")
-
-    def speech_to_text(self) -> str:
-        """"""
-        try:
-            with sr.Microphone() as mic:
-                print("Arni is listening...")
-                self.listener.adjust_for_ambient_noise(mic)
-                utterance = self.listener.listen(mic, timeout=self.timeout)
-                content = self.listener.recognize_google(utterance)
-                return content
-        except (sr.UnknownValueError, sr.RequestError, sr.WaitTimeoutError):
-            return ""
-
-    def text_to_speech(self, text: str):
-        print(f"Saying: {text}")
-        cmd = f'gtts-cli --nocheck "{text}" | mpg123 -q -'
-        print(cmd)
-        os.system(cmd)
-
-
-
-def main():
-    print("Creating virtual assistant...")
-    va = VirtualAssistant(actions=[BrowserAction(),EchoAction(), StopAction()])
-    # breakpoint()
-    while True:
-        content = va.speech_to_text()
-        va.parse_action(content)
-
-if __name__ == "__main__":
-    main()