voice_to_text_systemdev -checkpoint-checkpoint.py

#!/usr/bin/env python
# coding: utf-8

# In[29]:


import ipywidgets as widgets
from IPython.display import display, clear_output
from threading import Thread
from queue import Queue
import time

messages = Queue()
recordings = Queue()

record_button = widgets.Button(
    description="Record",
    disabled=False,
    button_style="success",
    icon="microphone"
)

stop_button = widgets.Button(
    description="Stop",
    disabled=False,
    button_style="warning",
    icon="stop"
)

output = widgets.Output()

def record_microphone():
    while not messages.empty():
        time.sleep(1)  # Simulate recording
        recordings.put("Audio recorded.")  # Simulated recorded audio data

def speech_recognition(output_widget):
    while not messages.empty():
        time.sleep(2)  # Simulate transcription
        with output_widget:
            clear_output(wait=True)
            display("Transcription: Hello, how are you?")  # Simulated transcription result

def start_recording(data):
    if not messages.empty():
        return  # Recording already in progress

    messages.put(True)
    with output:
        clear_output(wait=True)
        display("Starting...")

    record = Thread(target=record_microphone)
    record.start()

    transcribe = Thread(target=speech_recognition, args=(output,))
    transcribe.start()

def stop_recording(data):
    if messages.empty():
        return  # No recording in progress

    messages.get()
    with output:
        clear_output(wait=True)
        display("Stopped.")

record_button.on_click(start_recording)
stop_button.on_click(stop_recording)

display(widgets.HBox([record_button, stop_button]), output)


# In[30]:


get_ipython().system('python -m pip install pyaudio')


# In[31]:


import pyaudio

p = pyaudio.PyAudio()
for i in range(p.get_device_count()):
    print(p.get_device_info_by_index(i))

p.terminate()


# In[32]:


import pyaudio
from queue import Queue

CHANNELS = 1
FRAME_RATE = 16000
RECORD_SECONDS = 20
AUDIO_FORMAT = pyaudio.paInt16
SAMPLE_SIZE = 2

messages = Queue()
recordings = Queue()

def record_microphone(chunk=1024):
    p = pyaudio.PyAudio()

    stream = p.open(format=AUDIO_FORMAT,
                    channels=CHANNELS,
                    rate=FRAME_RATE,
                    input=True,
                    input_device_index=1,
                    frames_per_buffer=chunk)

    frames = []

    while not messages.empty():
        data = stream.read(chunk)
        frames.append(data)

        if len(frames) >= int(FRAME_RATE * RECORD_SECONDS / chunk):
            recordings.put(frames.copy())
            frames = []

    stream.stop_stream()
    stream.close()
    p.terminate()


# In[33]:


import subprocess
import json
from vosk import Model, KaldiRecognizer

model = Model(model_name="vosk-model-en-us-0.42-gigaspeech")
rec = KaldiRecognizer(model, FRAME_RATE)
rec.SetWords(True)

def speech_recognition(output):
    while not messages.empty():
        frames = recordings.get()

        rec.AcceptWaveform(b''.join(frames))
        result = rec.Result()
        text = json.loads(result)["text"]

        cased = subprocess.check_output("python recasepunc/recasepunc.py predict recasepunc/checkpoint", shell=True, text=True, input=text)
        output.append_stdout(cased)


# In[2]:


def my_function(input1, input2):
    # Process the inputs and generate the output
    output = f"Processed {input1} and {input2}"
    return output


# In[5]:


import gradio as gr

# Define the function you want to expose through Gradio
def my_function(input1, input2):
    output = f"Processed {input1} and {input2}"
    return output

# Create the Gradio interface
iface = gr.Interface(
    fn=my_function,
    inputs=[gr.Textbox(label="Input 1"), gr.Textbox(label="Input 2")],
    outputs=gr.Textbox(label="Output")
)

# Launch the interface with a public link
iface.launch(share=True)


# In[ ]: