polly.sh

#!/bin/bash

# Define default voice
default_voice="Ruth"

# Define available voices
voices=(
    "Ruth:Female English (US)"
    "Matthew:Male English (US)"
    "Amy:Female English (British)"
)

# Function to display usage
usage() {
    echo "Usage: $0 [-v <voice>]" >&2
    echo "Available voices:" >&2
    for voice in "${voices[@]}"; do
        name="${voice%:*}"
        desc="${voice#*:}"
        if [[ "$name" == "$default_voice" ]]; then
            echo "  $name: $desc - Default" >&2
        else
            echo "  $name: $desc" >&2
        fi
    done
    exit 1
}

# Check if AWS CLI is installed
if ! command -v aws &> /dev/null; then
    echo "AWS CLI is not installed. Please install it first."
    exit 1
fi

# Parse command-line options
while getopts ":v:" opt; do
    case $opt in
        v)
            voice=$OPTARG
            valid=0
            for v in "${voices[@]}"; do
                if [[ "${v%:*}" == "$voice" ]]; then
                    valid=1
                    break
                fi
            done
            if [ $valid -eq 0 ]; then
                echo "Invalid voice: $voice" >&2
                usage
            fi
            ;;
        *)
            echo "Invalid option: -$OPTARG" >&2
            usage
            ;;
    esac
done
shift $((OPTIND - 1))

# Set default voice if not provided
voice=${voice:-$default_voice}

# Check if input is from a pipe or keyboard
if [ -t 0 ]; then
    echo "Enter text to synthesize (press Ctrl+D to finish):"
    text=$(cat)
else
    text=$(cat)
fi

# Create temporary files
tmp_text=$(mktemp)
tmp_voice=$(mktemp)

# Write text to temporary file
echo "$text" > "$tmp_text"

# Synthesize speech using AWS Polly
echo "Using ${voice} voice."
aws polly synthesize-speech --output-format mp3 --region us-east-1 \
    --text "file://$tmp_text" \
    --voice-id "$voice" --engine neural "$tmp_voice"

# Play the synthesized speech
afplay "$tmp_voice"

# Clean up temporary files
rm "$tmp_text" "$tmp_voice"