python code voice to text

1  import speech_recognition as sr
2
3
4        def main():
5
6            r = sr.Recognizer()
7
8            with sr.Microphone() as source:
9                r.adjust_for_ambient_noise(source)
10
11                audio = r.listen(source)
12
13                try:
14
15                    print(r.recognize_google(audio))
16
17                except Exception as e:
18                    print("Error :  " + str(e))
19
20
21                with open("recorded.wav", "wb") as f:
22                    f.write(audio.get_wav_data())
23
24
25        if __name__ == "__main__":
26            main()

1# importing libraries 
2import speech_recognition as sr 
3import os 
4from pydub import AudioSegment
5from pydub.silence import split_on_silence
6
7# create a speech recognition object
8r = sr.Recognizer()
9
10# a function that splits the audio file into chunks
11# and applies speech recognition
12def get_large_audio_transcription(path):
13    """
14    Splitting the large audio file into chunks
15    and apply speech recognition on each of these chunks
16    """
17    # open the audio file using pydub
18    sound = AudioSegment.from_wav(path)  
19    # split audio sound where silence is 700 miliseconds or more and get chunks
20    chunks = split_on_silence(sound,
21        # experiment with this value for your target audio file
22        min_silence_len = 500,
23        # adjust this per requirement
24        silence_thresh = sound.dBFS-14,
25        # keep the silence for 1 second, adjustable as well
26        keep_silence=500,
27    )
28    folder_name = "audio-chunks"
29    # create a directory to store the audio chunks
30    if not os.path.isdir(folder_name):
31        os.mkdir(folder_name)
32    whole_text = ""
33    # process each chunk 
34    for i, audio_chunk in enumerate(chunks, start=1):
35        # export audio chunk and save it in
36        # the `folder_name` directory.
37        chunk_filename = os.path.join(folder_name, f"chunk{i}.wav")
38        audio_chunk.export(chunk_filename, format="wav")
39        # recognize the chunk
40        with sr.AudioFile(chunk_filename) as source:
41            audio_listened = r.record(source)
42            # try converting it to text
43            try:
44                text = r.recognize_google(audio_listened)
45            except sr.UnknownValueError as e:
46                print("Error:", str(e))
47            else:
48                text = f"{text.capitalize()}. "
49                print(chunk_filename, ":", text)
50                whole_text += text
51    # return the text for all chunks detected
52    return whole_text