Memanipulasi file audio dengan PyDub
Pemrosesan Bahasa Lisan dengan Python
Daniel Bourke
Machine Learning Engineer/YouTube Creator
Menurunkannya ke 11
# Impor file audio
wav_file = AudioSegment.from_file("wav_file.wav")
# Kurangi 60 dB
quiet_wav_file = wav_file - 60
# Coba kenali audio yang pelan
recognizer.recognize_google(quiet_wav_file)
UnknownValueError:
Menaikkan volume
# Tambah volume 10 dB
louder_wav_file = wav_file + 10
# Coba kenali
recognizer.recognize_google(louder_wav_file)
this is a wav file
Semua terdengar sama
# Impor AudioSegment dan normalisasi
from pydub import AudioSegment
from pydub.effects import normalize
from pydub.playback import play
# Impor audio dengan volume tidak rata
loud_quiet = AudioSegment.from_file("loud_quiet.wav")
# Normalisasi level suara
normalized_loud_quiet = normalize(loud_quiet)
# Cek suaranya
play(normalized_loud_quiet)
Remix file audio Anda
# Impor audio dengan statis di awal
static_at_start = AudioSegment.from_file("static_at_start.wav")
# Hapus statis dengan slicing
no_static_at_start = static_at_start[5000:]
# Cek suara baru
play(no_static_at_start)
Remix file audio Anda
# Impor dua file audio
wav_file_1 = AudioSegment.from_file("wav_file_1.wav")
wav_file_2 = AudioSegment.from_file("wav_file_2.wav")
# Gabungkan dua file audio
wav_file_3 = wav_file_1 + wav_file_2
# Cek suaranya
play(wav_file_3)
# Gabungkan dua file wav lalu buat lebih keras
louder_wav_file_3 = wav_file_1 + wav_file_2 + 10
Memisahkan audio Anda
# Impor audio panggilan telepon
phone_call = AudioSegment.from_file("phone_call.wav")
# Cari jumlah kanal
phone_call.channels
2
# Pisah stereo ke mono
phone_call_channels = phone_call.split_to_mono()
phone_call_channels
[<pydub.audio_segment.AudioSegment, <pydub.audio_segment.AudioSegment>]
Memisahkan audio Anda
# Cari jumlah kanal dari item pertama
phone_call_channels[0].channels
1
# Kenali kanal pertama
recognizer.recognize_google(phone_call_channel_1)
the pydub library is really useful
Ayo berlatih!
Pemrosesan Bahasa Lisan dengan Python
