Spaces:
Running
Running
File size: 2,152 Bytes
1d8403e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 | # ==================================================================================================
# DEEPFAKE AUDIO - encoder/params_data.py (Acoustic Feature Hyperparameters)
# ==================================================================================================
#
# 📝 DESCRIPTION
# This configuration module defines the signal processing constants for the
# speaker encoder. It standardizes window lengths, sampling rates, and VAD
# sensitivities, ensuring consistency between training data preparation and
# real-time inference.
#
# 👤 AUTHORS
# - Amey Thakur (https://github.com/Amey-Thakur)
# - Mega Satish (https://github.com/msatmod)
#
# 🤝🏻 CREDITS
# Original Real-Time Voice Cloning methodology by CorentinJ
# Repository: https://github.com/CorentinJ-Real-Time-Voice-Cloning
#
# 🔗 PROJECT LINKS
# Repository: https://github.com/Amey-Thakur/DEEPFAKE-AUDIO
# Video Demo: https://youtu.be/i3wnBcbHDbs
# Research: https://github.com/Amey-Thakur/DEEPFAKE-AUDIO/blob/main/DEEPFAKE-AUDIO.ipynb
#
# 📜 LICENSE
# Released under the MIT License
# Release Date: 2021-02-06
# ==================================================================================================
# --- MEL-FILTERBANK CONFIGURATION ---
mel_window_length = 25 # Spectral analysis window (ms)
mel_window_step = 10 # Temporal stride between windows (ms)
mel_n_channels = 40 # Number of mel-scale frequency bins
# --- AUDIO TEMPORAL RESOLUTION ---
sampling_rate = 16000 # Global acoustic sampling frequency (Hz)
partials_n_frames = 160 # Sequence length for training utterances (1.6s)
inference_n_frames = 80 # Minimal sequence length for identity derivation (0.8s)
# --- VOICE ACTIVITY DETECTION (VAD) ---
# Sensitivity parameters for distinguishing speech from silence.
vad_window_length = 30 # Temporal resolution of VAD decisions (ms)
vad_moving_average_width = 8 # Smoothing factor for binary speech decisions
vad_max_silence_length = 6 # Maximum allowed internal silence gap before segmentation
# --- AMPLITUDE NORMALIZATION ---
audio_norm_target_dBFS = -30 # Target spectral energy level in Decibels
|