speech-articulatory-coding 0.1.0

Python code for analyzing and synthesizing articulatory features of speech

Speech Articulatory Coding (SPARC)

Paper | Audio Samples | Colab Demo

This is the official code base for Coding Speech through Vocal Tract Kinematics.

Installation

git clone https://github.com/cheoljun95/Speech-Articulatory-Coding.git
cd Speech-Articulatory-Coding
pip install -e .

Usage

Load Model

from sparc import load_model
coder = load_model("en", device= "cpu")     # For using CPU
coder = load_model("en", device= "cuda:0")  # For using GPU

For pitch tracker, we found PENN is fast at inference. You can activate that with use_penn=True. The default is using torchcrepe.

coder = load_model("en", device= "cpu", use_penn=True)    # Use PENN for pitch tracker

For inversion only, you can use the following,

coder = load_model("feature_extraction") 
coder_from_config = load_model(config="configs/feature_extraction.yaml")

The following model checkpoints are offered. You can replace en with other models (multi or en+) in load_model.

Model	Language	Training Dataset
en	English	LibriTTS-R
multi	Multi	LibriTTS-R, Multilignual LibriSpeech, AISHELL, JVS, KSS
en+	English	LibriTTS-R, LibriTTS, EXPRESSO

Articulatory Analysis

code = coder.encode(WAV_FILE)          # Single inference
codes = coder.encode([WAV_FILE1, WAV_FILE2, ...]) # Batched processing

The articulatory code outputs have the following format.

# All features are in 50 Hz except speaker encoding
{"ema": (L, 12) array, #'TDX','TDY','TBX','TBY','TTX','TTY','LIX','LIY','ULX','ULY','LLX','LLY'
 "loudness": (L, 1) array, 
 "pitch": (L, 1) array, 
 "periodicity": (L, 1) array, # auxiliary output of pitch tracker
 "pitch_stats": (pitch mean, pitch std),
 "spk_emb": (spk_emb_dim,) array, # all shared models use spk_emb_dim=64
 "ft_len": Length of features, # usefull when batched processing with padding
}

Articulatory Synthesis

wav = coder.decode(**code)
sr = coder.sr

Voice Conversion

wav = coder.convert(SOURCE_WAV_FILE, TARGET_WAV_FILE)
sr = coder.sr

Demo

Please check notebooks/demo.ipynb for a demonstration of the functions.

Training

Feature extraction

Check scripts/encode_audio.py and an example script for extracting LibriTTS, scripts/extract_libritts.sh

TODO

• Add training codes.
• Add pypi installation.