renatoxsr/epub-tts
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Unified aedocw backends

Browse Source
This commit is contained in:
renatoxsr
2026-08-07 00:03:19 -03:00
parent da8940bdd4
commit b1c7a42172
12 changed files with 169 additions and 397 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.gitignore
+4 -1
View File
@@ -137,6 +137,8 @@ ENV/
env.bak/
venv.bak/
.tmp-samples/
.tmp*
.venv*
# Spyder project settings
.spyderproject
@@ -223,7 +225,8 @@ $RECYCLE.BIN/
.LSOverride
# Icon must end with two \r
Icon␍
Icon
# Thumbnails
._*
.python-version
+1
View File
@@ -0,0 +1 @@
3.11
README.md
+29
View File
@@ -2,6 +2,35 @@
Python project to gather several projects that convert EPUB files to audiobooks (M4B files, mostly), with different TTS backends.
# TL;DR
## Setup
```sh
git clone https://git.silveirarosa.com/renatoxsr/epub-tts
cd .\epub-tts\
pyenv local 3.11
python3 -m venv .venv
git submodule update --init --recursive
```
## Install on Windows
```pwsh
.\.venv\Scripts\python.exe -m pip install -e '.[dev]'
.\.venv\Scripts\download-epub-minimal.exe
.\.venv\Scripts\download-moby-dick.exe
.\.venv\Scripts\download-moby-dick-media-overlays.exe
.\.venv\Scripts\epub-tts.exe -i .\.tmp-samples\minimal.epub -o .\.tmp-m4b\minimal.m4b -b epub2tts-edge
```
## Install on POSIX
```sh
.\.venv\bin\python -m pip install -e '.[dev]'
.\.venv\bin\download-epub-minimal
.\.venv\bin\download-moby-dick
.\.venv\bin\download-moby-dick-media-overlays
.\.venv\bin\epub-tts -i ./.tmp-samples/minimal.epub -o ./.tmp-m4b/minimal.m4b -b epub2tts-edge
```
# Requirements
Recent Python (>=3.11), pip (>=22.3), some sort of virtula environment recommended (such as the builtin venv package), and a Rust compatible platform (avoid embedded systems which do not have pre-built binaries for the compiled requirements, e.g., Android's Termux try to build everything from sources and fails miserably when doing so).
src/epub_tts/__main__.py
+19 -6
View File
@@ -23,7 +23,9 @@ optional arguments:
-b {default,edge,epub2tts,epub2tts-edge,epub2tts-chatterbox,epub2tts-kokoro}, --backend {default,edge,epub2tts,epub2tts-edge,epub2tts-chatterbox,epub2tts-kokoro}
Backend to use for TTS
"""
import os
import sys
from pathlib import Path
from argparse import ArgumentParser
from . import tts_aedocw, tts_edge, tts_generic
@@ -39,6 +41,7 @@ backend = {
"epub2tts-edge": tts_aedocw.AedocwEpub2TTSEdge,
"epub2tts-chatterbox": tts_aedocw.AedocwChatterbox,
"epub2tts-kokoro": tts_aedocw.AedocwKokoro,
"generic-epub2tts": tts_aedocw.Epub2TTS
}
@@ -63,18 +66,28 @@ def main(args=None):
)
args = p.parse_args(args or sys.argv[1:])
print(
f"Converting {args.input} to {args.output} using voice "
f"{args.voice}, language {args.language}, "
f"backend {args.backend}, ..."
f"Converting {args.input} to {args.output} "
f"using {args.backend} backend "
"and args: " + ", ".join(
f"{k}={v}" for k, v in vars(args).items()
if k not in ("input", "output", "backend") and v is not None)
)
input_path = Path(args.input).resolve()
if not input_path.exists():
raise FileNotFoundError(f"Input file {input_path} does not exist.")
output_path = Path(args.output).resolve()
if not output_path.parent.exists():
os.makedirs(output_path.parent, exist_ok=True)
backend_cls = backend[args.backend]
backend_instance = backend_cls(
backend_cmd=None,
language=args.language,
voice=args.voice,
)
backend_instance.run(args.input, args.output)
backend_instance.run(
input_path, output_path,
**vars(args),
)
if __name__ == "__main__":
src/epub_tts/tts_aedocw.py
+25 -94
View File
@@ -13,12 +13,6 @@ The vendored packages are run in isolated virtual environments when the
corresponding console script is not available in the current environment.
"""
from typing import Optional
from . import tts_epub2tts as epub2tts_adapter
from . import tts_epub2tts_chatterbox as epub2tts_chatterbox_adapter
from . import tts_epub2tts_edge as epub2tts_edge_adapter
from . import tts_epub2tts_kokoro as epub2tts_kokoro_adapter
from .tts_generic import GenericTTSBackend
@@ -35,110 +29,45 @@ class AedocwBackend(GenericTTSBackend):
but the vendored adapters may still run the package in their own venv if
the console script is not present.
"""
_CMD_MAP = {
"epub2tts": "epub2tts",
"epub2tts-edge": "epub2tts-edge",
"epub2tts-chatterbox": "epub2tts-chatterbox",
"epub2tts-kokoro": "epub2tts-kokoro",
}
_RUNNER_MAP = {
"epub2tts": epub2tts_adapter.run,
"epub2tts-edge": epub2tts_edge_adapter.run,
"epub2tts-chatterbox": epub2tts_chatterbox_adapter.run,
"epub2tts-kokoro": epub2tts_kokoro_adapter.run,
}
def __init__(
self,
repo: str = "epub2tts",
backend_cmd: Optional[str] = None,
language: Optional[str] = None,
voice: Optional[str] = None,
):
if backend_cmd is None:
backend_cmd = self._CMD_MAP.get(repo, "epub2tts")
super().__init__(backend_cmd=backend_cmd, language=language, voice=voice)
self.repo = repo
def run(self, input_source, output_dest):
runner = self._RUNNER_MAP.get(self.repo)
if runner is None:
raise ValueError(f"Unknown repository backend: {self.repo}")
return runner(
input_source,
output_dest,
language=self.language,
voice=self.voice,
backend_cmd=self.backend_cmd,
)
FLAGS = {
"speaker": "--speaker",
"voice": "--speaker",
"language": "--language",
}
INTERMEDIATE_TXT = True
class AedocwEpub2TTS(AedocwBackend):
"""Backend configured for github.com/aedocw/epub2tts."""
def __init__(
self,
backend_cmd: Optional[str] = None,
language: Optional[str] = None,
voice: Optional[str] = None,
):
super().__init__(
repo="epub2tts", backend_cmd=backend_cmd, language=language, voice=voice
)
pass
class AedocwEpub2TTSEdge(AedocwBackend):
"""Backend configured for github.com/aedocw/epub2tts-edge."""
DEFAULT_SPEAKER = "en-US-AndrewNeural"
def __init__(
self,
backend_cmd: Optional[str] = None,
language: Optional[str] = None,
voice: Optional[str] = None,
):
super().__init__(
repo="epub2tts-edge",
backend_cmd=backend_cmd,
language=language,
voice=voice,
)
class TTSEdge(AedocwEpub2TTSEdge):
"""Backward-compatible edge backend wrapper."""
pass
class AedocwChatterbox(AedocwBackend):
"""Backend configured for github.com/aedocw/epub2tts-chatterbox."""
def __init__(
self,
backend_cmd: Optional[str] = None,
language: Optional[str] = None,
voice: Optional[str] = None,
):
super().__init__(
repo="epub2tts-chatterbox",
backend_cmd=backend_cmd,
language=language,
voice=voice,
)
DEFAULT_SAMPLE = "none"
class AedocwKokoro(AedocwBackend):
"""Backend configured for github.com/aedocw/epub2tts-kokoro."""
DEFAULT_SPEAKER = "af_heart"
def __init__(
self,
backend_cmd: Optional[str] = None,
language: Optional[str] = None,
voice: Optional[str] = None,
):
super().__init__(
repo="epub2tts-kokoro",
backend_cmd=backend_cmd,
language=language,
voice=voice,
)
class Epub2TTS(AedocwBackend):
"""Backward-compatible chatterbox backend wrapper."""
pass
__all__ = [
@@ -147,4 +76,6 @@ __all__ = [
"AedocwEpub2TTSEdge",
"AedocwChatterbox",
"AedocwKokoro",
"TTSEdge",
"Epub2TTS",
]
src/epub_tts/tts_edge.py
-14
View File
@@ -1,14 +0,0 @@
#!/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)
# Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
"""Compatibility shim for the edge backend in epub-tts."""
from .tts_aedocw import AedocwEpub2TTSEdge
class TTSEdge(AedocwEpub2TTSEdge):
"""Backward-compatible edge backend wrapper."""
pass
src/epub_tts/tts_epub2tts.py
-75
View File
@@ -1,75 +0,0 @@
#!/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)
# Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
"""Adapter for the vendored github.com/aedocw/epub2tts repository."""
import subprocess
from pathlib import Path
from typing import Optional
from .vendor_utils import build_run_command
SUPPORTED_AUDIO_FORMATS = {
".m4b": "m4b",
".wav": "wav",
".flac": "flac",
}
def _normalize_path(path_input: Path | str) -> Path:
return Path(path_input)
def _build_command(
input_source: Path,
output_dest: Path,
language: Optional[str] = None,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> list[str]:
command = build_run_command("epub2tts", "epub2tts", backend_cmd=backend_cmd)
command.append(str(input_source))
if language:
command.extend(["--language", language])
if voice:
command.extend(["--speaker", voice])
output_format = SUPPORTED_AUDIO_FORMATS.get(output_dest.suffix.lower(), "m4b")
if output_format != "m4b":
command.extend(["--audioformat", output_format])
return command
def run(
input_source: Path | str,
output_dest: Path | str,
language: Optional[str] = None,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> subprocess.CompletedProcess:
input_path = _normalize_path(input_source)
output_path = _normalize_path(output_dest)
output_path.parent.mkdir(parents=True, exist_ok=True)
command = _build_command(
input_path,
output_path,
language=language,
voice=voice,
backend_cmd=backend_cmd,
)
completed = subprocess.run(command, cwd=str(input_path.parent), check=True)
generated_path = input_path.with_suffix(
SUPPORTED_AUDIO_FORMATS.get(output_path.suffix.lower(), "m4b")
)
if generated_path != output_path:
generated_path.replace(output_path)
return completed
src/epub_tts/tts_epub2tts_chatterbox.py
-62
View File
@@ -1,62 +0,0 @@
#!/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)
# Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
"""Adapter for the vendored github.com/aedocw/epub2tts-chatterbox repository."""
import subprocess
from pathlib import Path
from typing import Optional
from .vendor_utils import build_run_command
DEFAULT_SAMPLE = "none"
def _normalize_path(path_input: Path | str) -> Path:
return Path(path_input)
def _build_command(
input_source: Path,
sample: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> list[str]:
command = build_run_command(
"epub2tts-chatterbox",
"epub2tts_chatterbox",
backend_cmd=backend_cmd,
)
command.append(str(input_source))
if sample and sample != DEFAULT_SAMPLE:
command.extend(["--sample", sample])
return command
def run(
input_source: Path | str,
output_dest: Path | str,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> subprocess.CompletedProcess:
input_path = _normalize_path(input_source)
output_path = _normalize_path(output_dest)
output_path.parent.mkdir(parents=True, exist_ok=True)
sample_text = DEFAULT_SAMPLE
cwd = input_path.parent
if voice:
voice_path = Path(voice)
sample_text = voice_path.name
if voice_path.parent != Path(""):
cwd = str(voice_path.parent)
command = _build_command(input_path, sample=sample_text, backend_cmd=backend_cmd)
completed = subprocess.run(command, cwd=cwd, check=True)
generated_path = Path(cwd) / f"{input_path.stem} ({sample_text}).m4b"
if generated_path != output_path:
generated_path.replace(output_path)
return completed
src/epub_tts/tts_epub2tts_edge.py
-53
View File
@@ -1,53 +0,0 @@
#!/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)
# Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
"""Adapter for the vendored github.com/aedocw/epub2tts-edge repository."""
import subprocess
from pathlib import Path
from typing import Optional
from .vendor_utils import build_run_command
DEFAULT_SPEAKER = "en-US-AndrewNeural"
def _normalize_path(path_input: Path | str) -> Path:
return Path(path_input)
def _build_command(
input_source: Path,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> list[str]:
command = build_run_command(
"epub2tts-edge", "epub2tts_edge", backend_cmd=backend_cmd
)
command.append(str(input_source))
if voice:
command.extend(["--speaker", voice])
return command
def run(
input_source: Path | str,
output_dest: Path | str,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> subprocess.CompletedProcess:
input_path = _normalize_path(input_source)
output_path = _normalize_path(output_dest)
output_path.parent.mkdir(parents=True, exist_ok=True)
speaker = voice or DEFAULT_SPEAKER
command = _build_command(input_path, voice=speaker, backend_cmd=backend_cmd)
completed = subprocess.run(command, cwd=str(input_path.parent), check=True)
generated_path = input_path.parent / f"{input_path.stem} ({speaker}).m4b"
if generated_path != output_path:
generated_path.replace(output_path)
return completed
src/epub_tts/tts_epub2tts_kokoro.py
-53
View File
@@ -1,53 +0,0 @@
#!/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)
# Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
"""Adapter for the vendored github.com/aedocw/epub2tts-kokoro repository."""
import subprocess
from pathlib import Path
from typing import Optional
from .vendor_utils import build_run_command
DEFAULT_SPEAKER = "af_heart"
def _normalize_path(path_input: Path | str) -> Path:
return Path(path_input)
def _build_command(
input_source: Path,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> list[str]:
command = build_run_command(
"epub2tts-kokoro", "epub2tts_kokoro", backend_cmd=backend_cmd
)
command.append(str(input_source))
if voice:
command.extend(["--speaker", voice])
return command
def run(
input_source: Path | str,
output_dest: Path | str,
voice: Optional[str] = None,
backend_cmd: Optional[str] = None,
) -> subprocess.CompletedProcess:
input_path = _normalize_path(input_source)
output_path = _normalize_path(output_dest)
output_path.parent.mkdir(parents=True, exist_ok=True)
speaker = voice or DEFAULT_SPEAKER
command = _build_command(input_path, voice=speaker, backend_cmd=backend_cmd)
completed = subprocess.run(command, cwd=str(input_path.parent), check=True)
generated_path = input_path.parent / f"{input_path.stem} ({speaker}).m4b"
if generated_path != output_path:
generated_path.replace(output_path)
return completed
src/epub_tts/tts_generic.py
+84 -38
View File
@@ -8,16 +8,23 @@
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.
gen_input_flag(input_path: Path) -> str: Return backend flag for input file or directory.
gen_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.
"""
@@ -28,52 +35,91 @@ from typing import Optional, Union
PathLike = Union[str, Path]
from .vendor_utils import build_run_command
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
"""Generic TTS backend for handling text-to-speech conversion.
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.
"""
FLAGS = {}
INTERMEDIATE_TXT = False
DEFAULT_SPEAKER = None
DEFAULT_SAMPLE = None
SUPPORTED_AUDIO_FORMATS = {
# subclasses may override this
".m4b": "m4b",
".wav": "wav",
".flac": "flac",
".mp3": "mp3",
".ogg": "ogg",
}
_INPUT_FLAG_MAP = lambda x: [
"--input-dir", x] if x else ["--input-file", x]
_OUTPUT_FLAG_MAP = lambda x: [
"--output-dir", x] if x else ["--output-file", x]
def __init__(self,**kwargs):
for k,v in kwargs.items():
setattr(self, k, v)
def _normalize_path(self, value: PathLike) -> Path:
if isinstance(value, Path):
return value
return Path(value)
return value.resolve()
else:
return Path(value).resolve()
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 gen_input_flag(self, input_path: Path) -> str:
return self._INPUT_FLAG_MAP(self._normalize_path(input_path).is_dir())
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),
]
def gen_output_flag(self, output_path: Path) -> str:
return self._OUTPUT_FLAG_MAP(self._normalize_path(output_path).is_dir())
if self.language:
command.extend(["--language", self.language])
if self.voice:
command.extend(["--voice", self.voice])
def _build_command(self, input_source: PathLike, output_dest: PathLike) -> list[str]:
command = [self.backend_cmd]
command.extend(self.gen_input_flag(input_source))
command.extend(self.gen_output_flag(output_dest))
for flag_name, var_name in self.FLAGS.items():
if getattr(self, flag_name, None) is not None:
command.extend([
f"--{flag_name.replace('_', '-')}",
getattr(self, var_name)])
return command
def run(self,
input_source: PathLike,
output_dest: PathLike,
**kwargs
) -> subprocess.CompletedProcess:
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)
completed = subprocess.run(
self._build_command(input_source, output_dest),
cwd=str(self._normalize_path(input_source).parent),
check=True)
if self.INTERMEDIATE_TXT:
txt_file = self._normalize_path(input_source).with_suffix(".txt")
if txt_file.exists():
completed = subprocess.run(
self._build_command(txt_file, output_dest),
cwd=str(self._normalize_path(input_source).parent),
check=True
)
return completed
src/epub_tts/vendor_utils.py
+7 -1
View File
@@ -64,4 +64,10 @@ def build_run_command(
if backend_cmd is None or backend_cmd == repo_name:
python_executable = ensure_venv(repo_name)
return [str(python_executable), "-m", module_name]
return [backend_cmd]
else:
ensure_venv(repo_name)
venv_path = get_venv_path(repo_name)
return [str( venv_path / "Scripts" / f"{backend_cmd}.exe")
if sys.platform == "win32"
else str(venv_path / "bin" / backend_cmd)]