#!/bin/bash # Check if the correct number of arguments are provided if [ "$#" -ne 1 ]; then echo "Usage: $0 <speaker_name>" exit 1 fi # Assign the input parameter to a variable speaker_name="$1" # Set the path to the TTS model model_name="tts_models/multilingual/multi-dataset/xtts_v2" # Set the input text text="Hello World! This is a test for speech synthesis in english. Do you like it?" # Set the language index language_idx="en" # Set the CUDA usage flag use_cuda=1 # Set the output path out_path="/mnt/d/AiVoice/xtts/${speaker_name}.wav" # Run the TTS command tts --model_name "$model_name" --text "$text" --speaker_idx="$speaker_name" --language_idx="$language_idx" --use_cuda "$use_cuda" --out_path "$out_path" echo "Speech synthesis completed. Output saved to: $out_path"