### Install TensorTract2 via pip Source: https://github.com/altavo/tensortract2/blob/main/README.md Install the TensorTract2 package using pip. Run this command in your terminal to install the package. ```bash pip install tensortract2 ``` -------------------------------- ### Load TensorTract2 model with automatic weight download Source: https://github.com/altavo/tensortract2/blob/main/README.md Initializes the TensorTract2 model. On first initialization, weights are downloaded from Google Drive into the user's cache directory; wavlm-large is downloaded from Huggingface. Subsequent loads skip the download. ```python from tensortract2 import TensorTract2 # Load the model tt2 = TensorTract2() ``` -------------------------------- ### Load TensorTract2 weights manually with auto_load_weights=False Source: https://github.com/altavo/tensortract2/blob/main/README.md Use when you want to supply weights yourself instead of the default download. The model is always initialized in eval mode automatically, so no manual eval-mode call is needed. ```python tt2 = TensorTract2(auto_load_weights = False) tt2.load_weights( 'path/to/weights' ) ``` -------------------------------- ### Convert speech to articulatory parameters with speech_to_motor Source: https://github.com/altavo/tensortract2/blob/main/README.md Converts speech audio files to articulatory parameters. The input x can be a string or a list of strings; the output is a list of MotorSeries objects (see the target-approximation package). ```python # Load speech from an audio file and process it motor_data = tt2.speech_to_motor( x='path/to/audio.wav', ) ``` === COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.