speech to text v1.0.1

'''

Ethan Anderson
Balance of Nature
VtT v1.0.1
Oct. 20 2021

'''

# importing libraries 
import speech_recognition as sr 
import os
from pydub import AudioSegment
from pydub.silence import split_on_silence


#create a speech recognition object
r = sr.Recognizer()
#change this to the folder you have your wav files
VM_directory = r'vm temp wav'

def send_to_rr():
    print("Sending to R&R...")
    #put actual transfer process here
def send_to_customer_care():
    print("Sending to CC...")
    #put actual transfer process here
def send_to_new_sales():
    print("Sending to NS...")
    #put actual transfer process here
def send_to_customer_sales():
    print("Sending to CS...")
    #put actual transfer process here
def send_to_returns():
    print("Sending to Returns...")
    #put actual transfer process here
def send_to_declines():
    print("Sending to Declines...")
    #put actual transfer process here
def send_to_scheduling():
    print("Sending to Scheduling...")
    #put actual transfer process here
def no_matches():
    print("No Matches")
    #put actual transfer process here

all_departments_words = [
    {
        'words' : ["tracking", "shipment", "pending", "shipped", "shipping", "where my order is", "where is my order", "delay", "delayed", "not received", "not recieve"],
        'target' : send_to_rr,
        'message' : "Sent to Reception and Routing"
    },

    {
        'words' : ["stop", "discontinue", "backlog", "surplus", "postpone", "no shipments"],
        'target' : send_to_customer_care,
        'message' : "Sent to Customer Care"
    },

    {
        'words' : ["fox","new customer", "cost", "discount code", "commercial", "do not have a computer", "don't have a computer", "35%", "35 percent", "Fox"],
        'target' : send_to_new_sales,
        'message' : "Sent to New Sales"
    },

    {
        'words' : ["reinstate", "old account", "ordered before", "place an order", "place another order", "renew", "change in my order", "reactivate", "reorder", "make an order", "apple", "preferred customer"],
        'target' : send_to_customer_sales,
        'message' : "Sent to Customer Sales"
    },

    {
        'words' : ["refund", "money back", "charged", "RMA", "R M A", "refunded", "refunding", "refused", "refuse"],
        'target' : send_to_returns,
        'message' : "Sent to Returns"
    },

    {
        'words' : ["fraud", "update", "declined", "decline", "account is disabled", "account has been disabled", "declines", "updated"],
        'target' : send_to_declines,
        'message' : "Sent to Declines"
    },

    {
        'words' : ["appointment", "scheduled", "schedule", "reschedule", "coach"],
        'target' : send_to_scheduling,
        'message' : "Sent to Scheduling"
    },

    {
        'words' : ["return", "returns", "returning"],
        'target' : send_to_returns,
        'message' : "Sent to Returns"
    },

    {
        'words' : ["cancel", "cancelled", "cancelling", "canceling"],
        'target' : send_to_customer_care,
        'message' : "Sent to Customer Care"
    },

    {
        'words' : ["ship"],
        'target' : send_to_rr,
        'message' : "Sent to Reception and Routing"
    },

    {
        'words' : [' ', '-'],
        'target' : no_matches,
        'message' : "No Matches"
    }

]



####################################### FUNCTIONS ##########################################

# a function that splits the audio file into chunks
# and applies speech recognition - edited to make it one giant chunk instead of multiple small ones (would want to be split into small ones for conversations)
def get_large_audio_transcription(path):
    """
    Splitting the large audio file into chunks
    and apply speech recognition on each of these chunks
    """
    # open the audio file using pydub
    sound = AudioSegment.from_wav(path)  
    # split audio sound where silence is 700 miliseconds or more and get chunks
    chunks = split_on_silence(sound,
        # experiment with this value for your target audio file
        min_silence_len = 1500, #change this if you want to modify for conversations, detects the silence length to find breaks
        # adjust this per requirement
        silence_thresh = sound.dBFS-14,
        # keep the silence for 1 second, adjustable as well
        keep_silence=500,
    )
    folder_name = "audio-chunks"
    # create a directory to store the audio chunks
    if not os.path.isdir(folder_name):
        os.mkdir(folder_name)
    whole_text = ""
    # process each chunk 
    for i, audio_chunk in enumerate(chunks, start=1):
        # export audio chunk and save it in
        # the `folder_name` directory.
        chunk_filename = os.path.join(folder_name, f"chunk{i}.wav")
        audio_chunk.export(chunk_filename, format="wav")
        # recognize the chunk
        with sr.AudioFile(chunk_filename) as source:
            audio_listened = r.record(source)
            # try converting it to text
            try:
                text = r.recognize_google(audio_listened)
            except sr.UnknownValueError as e:
                print("Error:", str(e))
            else:
                text = f"{text.capitalize()}. "
                print(chunk_filename, ":", text)
                whole_text += text
    # return the text for all chunks detected
    return whole_text


def main():
    #checklist = []
    for filename in os.scandir(VM_directory):
        #only running through .wav files
        if filename.path.endswith(".wav"):
            vm_text_string = get_large_audio_transcription(filename)
            #flags for if there are no keywords
            flag = 0
            #loops through dictionaries
            for department in all_departments_words:
                if flag == 1:
                    break
                #loops through keywords
                for word in department['words']:
                    #loops words through VM and splits string into words
                    if word in vm_text_string:
                        #calls the function that will actually send the vm
                        department['target']()
                        #prints the "sent to ..." from dict
                        print(department['message'])
                        #checklist.append(department['message'])
                        flag += 1
                        #ends loop to prevent repeats
                        break
            #if no keywords are detected flag will stay 0
            if flag == 0:
                #runs function for no matches
                no_matches()
                #checklist.append('No Matches')
    '''print(checklist)
    answers = ['Sent to Customer Care', 'Sent to Customer Care', 'Sent to Customer Care', 'Sent to Customer Care', 'Sent to Customer Sales', 'Sent to Customer Sales', 'No Matches', 'No Matches', 'No Matches', 'Sent to New Sales', 'Sent to New Sales', 'Sent to Returns', 'Sent to Returns', 'Sent to Reception and Routing', 'Sent to Reception and Routing', 'Sent to Reception and Routing', 'Sent to Reception and Routing', 'Sent to Scheduling', 'Sent to Scheduling']
    i = -1
    while i < 19:
        i += 1
        if checklist[i] == answers[i]:
            print('correct')   
        else:
            print('incorrect')'''

main()