Changes to be committed:

new file:   AGENTS.md
	modified:   pyproject.toml
	modified:   src/epub_tts/__main__.py
	new file:   src/epub_tts/tts_backend.py

src/epub_tts/__main__.py

@@ -7,6 +7,15 @@
 import sys
 from argparse import ArgumentParser
 
+from epub_tts import tts_backend
+from epub_tts import tts_edge
+
+backend = {
+    # keys are the backend names, values are the corresponding TTS classes
+    "default": tts_backend.GenericTTSBackend,
+    "edge": tts_edge.TTSEdge,
+}
+
 
 def main(args=None):
     p = ArgumentParser(description="Convert EPUB to audio using TTS")
@@ -20,10 +29,15 @@ def main(args=None):
     p.add_argument("-o", "--output", help="Output audio file", required=True)
     p.add_argument("-l", "--language", help="Language to use for TTS")
     p.add_argument("-v", "--voice", help="Voice to use for TTS")
-
+    p.add_argument("-b", "--backend", 
+                   help="Backend to use for TTS",
+                   default="default",
+                   choices=backend.keys())
     args = p.parse_args(args or sys.argv[1:])
     print(
-        f"Converting {args.input} to {args.output} using voice {args.voice}, language {args.language}, ..."
+        f"Converting {args.input} to {args.output} using voice "
+        f"{args.voice}, language {args.language}, "
+        f"backend {args.backend}, ..."
     )
 
 

src/epub_tts/tts_backend.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# Copyright (c) 2025-2026 Renato Xavier da Silveira Rosa
+# See [LICENSE](./LICENSE) or [BSD-3-Clause-Clear](https://spdx.org/licenses/BSD-3-Clause-Clear.html)
+"""Generic TTS Backend for handling text-to-speech conversion.
+
+This module provides a generic TTS backend that can be used to convert text to speech using various TTS engines. It defines a `GenericTTSBackend` class that can be initialized with a specific backend command, language, and voice. The class provides methods to normalize paths, determine input and output flags based on the type of input/output (file or directory), build the command for the TTS engine, and run the command.
+
+Classes:
+    GenericTTSBackend:
+        Parameters:
+            backend_cmd (str): Command or executable for the backend.
+            language (Optional[str]): Language code for TTS output.
+            voice (Optional[str]): Voice name for TTS output.
+        Methods:
+            _normalize_path(value: PathLike) -> Path: Normalize a string or Path to Path.
+            _input_flag(input_path: Path) -> str: Return backend flag for input file or directory.
+            _output_flag(output_path: Path) -> str: Return backend flag for output file or directory.
+            _build_command(input_source: PathLike, output_dest: PathLike) -> list[str]: Build command list for subprocess.
+            run(input_source: PathLike, output_dest: PathLike) -> subprocess.CompletedProcess: Execute backend command.
+
+"""
+
+import subprocess
+from pathlib import Path
+from typing import Optional, Union
+
+PathLike = Union[str, Path]
+
+
+class GenericTTSBackend:
+    def __init__(
+        self,
+        backend_cmd: str,
+        language: Optional[str] = None,
+        voice: Optional[str] = None,
+    ):
+        self.backend_cmd = backend_cmd
+        self.language = language
+        self.voice = voice
+
+    def _normalize_path(self, value: PathLike) -> Path:
+        if isinstance(value, Path):
+            return value
+        return Path(value)
+
+    def _input_flag(self, input_path: Path) -> str:
+        return "--input-dir" if input_path.is_dir() else "--input-file"
+
+    def _output_flag(self, output_path: Path) -> str:
+        return "--output-dir" if output_path.is_dir() else "--output-file"
+
+    def _build_command(
+        self, input_source: PathLike, output_dest: PathLike
+    ) -> list[str]:
+        input_path = self._normalize_path(input_source)
+        output_path = self._normalize_path(output_dest)
+
+        command = [
+            self.backend_cmd,
+            self._input_flag(input_path),
+            str(input_path),
+            self._output_flag(output_path),
+            str(output_path),
+        ]
+
+        if self.language:
+            command.extend(["--language", self.language])
+        if self.voice:
+            command.extend(["--voice", self.voice])
+
+        return command
+
+    def run(
+        self, input_source: PathLike, output_dest: PathLike
+    ) -> subprocess.CompletedProcess:
+        command = self._build_command(input_source, output_dest)
+        return subprocess.run(command, check=True)