Audio Signal Comparison

Took two same speech audio signals(apple-like sound) and remove noise by using the Thresholding method.

Normalized both signals and create frame for both signals.

Applied various filters to remove remain noise and calculate Mel-frequency cepstral coefficients(MFCCs) of both signals.

Got 38% similarity by Cosine Similarity method.

Keywords : Audio Signal, Speech Processing, ASR, MFCC, Python3, Cosine Similarity.

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
.gitignore		.gitignore
Audio_similarity.py		Audio_similarity.py
LICENSE		LICENSE
README.md		README.md
audio_signal_similarity.ipynb		audio_signal_similarity.ipynb
audio_signal_similarity.py		audio_signal_similarity.py
client.ipynb		client.ipynb
cmd_client using speech and speak.ipynb		cmd_client using speech and speak.ipynb
feature _explore.ipynb		feature _explore.ipynb
feature.ipynb		feature.ipynb
pyaudio for ML model.ipynb		pyaudio for ML model.ipynb
speech.ipynb		speech.ipynb
speech_data.py		speech_data.py
speech_recognition.ipynb		speech_recognition.ipynb
wave0001.wav		wave0001.wav
wave0002.wav		wave0002.wav
wave001.wav		wave001.wav
wave002.wav		wave002.wav

Provide feedback