inference_melgan.sh

#!/bin/bash

set -euxo pipefail

if [ $# != 3 ]; then
  echo "Run inference on a trained MelGAN system"
  echo "Usage: $0 <cuda_id> <spk_id> <model_name>"
  echo "e.g.:"
  echo " $0 0 014 my_melgan_model"
  echo "<cuda_id> is the CUDA device you want to use."
  echo "<spk_id> is the speaker id from the wTIMIT corpus. Set it to 'all_spk' for speaker-indepedent training. "
  echo "<model_name> is the name of a directory where the trained model weights are located"
  exit 1
fi

CUDA_ID=$1
SPK_ID=$2
MODEL_NAME=$3

# Load model from checkpoint
LOAD_DIR=./data/checkpoint/${SPK_ID}/${MODEL_NAME}
# Save results
SAVE_DIR=./data/${MODEL_NAME}
# Path to audio input data
DATA_DIR=./data

echo "Run MelGAN inference"
export CUDA_VISIBLE_DEVICES=$CUDA_ID
python3 -m speech-conversion.melgan.inference \
  --save_path $SAVE_DIR \
  --load_path $LOAD_DIR \
  --data_path $DATA_DIR \
  --spk_id $SPK_ID

exit 0