test_s2s.sh

# audio
export DATASET_ID=test
export DIRECTION="enA-jaA"
export LINE_NO_START=0
export LINE_NO_END=10
export HF_DATASET="experiment-process-seamless-align"
python fetch_dataset_s2s.py
# tokenize
export DATASET_ID=test
export DIRECTION="enA-jaA"
export HF_DATASET="experiment-process-seamless-align"
python tokenize_dataset_s2s.py
# speaker embedding
export DATASET_ID=test
export DIRECTION="enA-jaA"
export HF_DATASET="experiment-process-seamless-align"
python attach_speaker_embedding_s2s.py