ttsforge 0.1.0__py3-none-any.whl

ttsforge/__init__.py

@@ -0,0 +1,114 @@
+"""
+ttsforge - Generate audiobooks from EPUB files with TTS.
+
+A CLI tool for converting EPUB books to audiobooks using Kokoro ONNX TTS.
+"""
+
+from pykokoro import GenerationConfig, KokoroPipeline, PipelineConfig
+from pykokoro.onnx_backend import (
+    VoiceBlend,
+    are_models_downloaded,
+    download_all_models,
+    download_model,
+    get_model_dir,
+)
+from pykokoro.tokenizer import (
+    EspeakConfig,
+    MAX_PHONEME_LENGTH,
+    Tokenizer,
+)
+from pykokoro.constants import SUPPORTED_LANGUAGES
+
+from .constants import (
+    DEFAULT_CONFIG,
+    LANGUAGE_DESCRIPTIONS,
+    SUPPORTED_OUTPUT_FORMATS,
+    VOICES,
+)
+
+# Import from pykokoro
+try:
+    from pykokoro.constants import SAMPLE_RATE
+    from pykokoro.onnx_backend import LANG_CODE_TO_ONNX
+except ImportError:
+    # Fallback values if pykokoro not installed
+    SAMPLE_RATE = 24000
+    LANG_CODE_TO_ONNX = {
+        "a": "en-us",
+        "b": "en-gb",
+        "e": "es",
+        "f": "fr-fr",
+        "h": "hi",
+        "i": "it",
+        "j": "ja",
+        "p": "pt",
+        "z": "zh",
+    }
+
+from .conversion import (
+    Chapter,
+    ConversionOptions,
+    ConversionProgress,
+    ConversionResult,
+    TTSConverter,
+)
+from .phonemes import (
+    FORMAT_VERSION,
+    PhonemeBook,
+    PhonemeChapter,
+    PhonemeSegment,
+    create_phoneme_book_from_chapters,
+    phonemize_text_list,
+)
+from .utils import (
+    load_config,
+    save_config,
+)
+from .cli.helpers import DEFAULT_SAMPLE_TEXT
+
+
+__all__ = [
+    # Constants
+    "DEFAULT_CONFIG",
+    "LANGUAGE_DESCRIPTIONS",
+    "SUPPORTED_OUTPUT_FORMATS",
+    "VOICES",
+    # Conversion
+    "Chapter",
+    "ConversionOptions",
+    "ConversionProgress",
+    "ConversionResult",
+    "TTSConverter",
+    # Pipeline (from pykokoro)
+    "GenerationConfig",
+    "KokoroPipeline",
+    "PipelineConfig",
+    "VoiceBlend",
+    "are_models_downloaded",
+    "download_all_models",
+    "download_model",
+    "get_model_dir",
+    # Tokenizer (from pykokoro)
+    "EspeakConfig",
+    "MAX_PHONEME_LENGTH",
+    "SAMPLE_RATE",
+    "SUPPORTED_LANGUAGES",
+    "Tokenizer",
+    # Phonemes
+    "FORMAT_VERSION",
+    "PhonemeBook",
+    "PhonemeChapter",
+    "PhonemeSegment",
+    "create_phoneme_book_from_chapters",
+    "phonemize_text_list",
+    # Utils
+    "load_config",
+    "save_config",
+    # herlpers
+    "DEFAULT_SAMPLE_TEXT",
+]
+
+try:
+    from ._version import version as __version__
+except ImportError:
+    __version__ = "0.0.0+unknown"

ttsforge/_version.py

@@ -0,0 +1,34 @@
+# file generated by setuptools-scm
+# don't change, don't track in version control
+
+__all__ = [
+    "__version__",
+    "__version_tuple__",
+    "version",
+    "version_tuple",
+    "__commit_id__",
+    "commit_id",
+]
+
+TYPE_CHECKING = False
+if TYPE_CHECKING:
+    from typing import Tuple
+    from typing import Union
+
+    VERSION_TUPLE = Tuple[Union[int, str], ...]
+    COMMIT_ID = Union[str, None]
+else:
+    VERSION_TUPLE = object
+    COMMIT_ID = object
+
+version: str
+__version__: str
+__version_tuple__: VERSION_TUPLE
+version_tuple: VERSION_TUPLE
+commit_id: COMMIT_ID
+__commit_id__: COMMIT_ID
+
+__version__ = version = '0.1.0'
+__version_tuple__ = version_tuple = (0, 1, 0)
+
+__commit_id__ = commit_id = None

ttsforge/audio_merge.py

@@ -0,0 +1,180 @@
+# ttsforge/audio_merge.py
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any, Protocol
+
+import numpy as np
+import soundfile as sf
+
+from .constants import SAMPLE_RATE
+from .utils import create_process, get_ffmpeg_path
+
+
+@dataclass(slots=True)
+class MergeMeta:
+    fmt: str
+    silence_between_chapters: float
+    title: str | None = None
+    author: str | None = None
+    cover_image: Path | None = None
+
+
+class AudioMerger:
+    class LogCallback(Protocol):
+        def __call__(self, message: str, level: str = "info") -> None: ...
+
+    def __init__(self, log: LogCallback):
+        self.log = log
+
+    def add_chapters_to_m4b(
+        self, output_path: Path, chapters: list[dict[str, Any]], cover: Path | None
+    ) -> None:
+        if len(chapters) <= 1:
+            return
+        ffmpeg = get_ffmpeg_path()
+
+        chapters_file = output_path.with_suffix(".chapters.txt")
+        chapters_file.write_text(self._ffmetadata(chapters), encoding="utf-8")
+
+        tmp_path = output_path.with_suffix(".tmp.m4b")
+        cmd = [
+            ffmpeg,
+            "-y",
+            "-i",
+            str(output_path),
+            "-i",
+            str(chapters_file),
+            "-map",
+            "0:a",
+            "-map_metadata",
+            "1",
+            "-map_chapters",
+            "1",
+            "-c:a",
+            "copy",
+        ]
+
+        if cover and cover.exists():
+            cmd += [
+                "-i",
+                str(cover),
+                "-map",
+                "2",
+                "-c:v",
+                "copy",
+                "-disposition:v",
+                "attached_pic",
+            ]
+
+        cmd.append(str(tmp_path))
+        proc = create_process(cmd, suppress_output=True)
+        rc = proc.wait()
+        if rc != 0:
+            raise RuntimeError("ffmpeg failed while adding m4b chapters")
+
+        tmp_path.replace(output_path)
+        chapters_file.unlink(missing_ok=True)
+
+    def merge_chapter_wavs(
+        self,
+        chapter_files: list[Path],
+        chapter_durations: list[float],
+        chapter_titles: list[str],
+        output_path: Path,
+        meta: MergeMeta,
+    ) -> None:
+        ffmpeg = get_ffmpeg_path()
+
+        concat_file = output_path.with_suffix(".concat.txt")
+        silence_file = output_path.parent / "_silence.wav"
+
+        if meta.silence_between_chapters > 0 and len(chapter_files) > 1:
+            self._write_silence_wav(silence_file, meta.silence_between_chapters)
+
+        with concat_file.open("w", encoding="utf-8") as f:
+            for i, ch in enumerate(chapter_files):
+                f.write(f"file '{ch.absolute()}'\n")
+                if i < len(chapter_files) - 1 and meta.silence_between_chapters > 0:
+                    f.write(f"file '{silence_file.absolute()}'\n")
+
+        cmd = [ffmpeg, "-y", "-f", "concat", "-safe", "0", "-i", str(concat_file)]
+
+        if meta.fmt == "m4b":
+            if meta.cover_image and meta.cover_image.exists():
+                cmd += [
+                    "-i",
+                    str(meta.cover_image),
+                    "-map",
+                    "0:a",
+                    "-map",
+                    "1",
+                    "-c:v",
+                    "copy",
+                    "-disposition:v",
+                    "attached_pic",
+                ]
+            cmd += [
+                "-c:a",
+                "aac",
+                "-q:a",
+                "2",
+                "-movflags",
+                "+faststart+use_metadata_tags",
+            ]
+            if meta.title:
+                cmd += ["-metadata", f"title={meta.title}"]
+            if meta.author:
+                cmd += ["-metadata", f"artist={meta.author}"]
+        elif meta.fmt == "opus":
+            cmd += ["-c:a", "libopus", "-b:a", "24000"]
+        elif meta.fmt == "mp3":
+            cmd += ["-c:a", "libmp3lame", "-q:a", "2"]
+        elif meta.fmt == "flac":
+            cmd += ["-c:a", "flac"]
+        elif meta.fmt == "wav":
+            cmd += ["-c:a", "pcm_s16le"]
+
+        cmd.append(str(output_path))
+        proc = create_process(cmd, suppress_output=True)
+        rc = proc.wait()
+        if rc != 0:
+            raise RuntimeError("ffmpeg failed while merging chapters")
+
+        concat_file.unlink(missing_ok=True)
+        silence_file.unlink(missing_ok=True)
+
+        if meta.fmt == "m4b" and len(chapter_files) > 1:
+            times = []
+            t = 0.0
+            for i, (dur, title) in enumerate(
+                zip(chapter_durations, chapter_titles, strict=False)
+            ):
+                times.append({"title": title, "start": t, "end": t + dur})
+                t += dur
+                if i < len(chapter_durations) - 1:
+                    t += meta.silence_between_chapters
+            self.add_chapters_to_m4b(output_path, times, meta.cover_image)
+
+    def _write_silence_wav(self, path: Path, duration: float) -> None:
+        samples = int(duration * SAMPLE_RATE)
+        audio = np.zeros(samples, dtype="float32")
+        with sf.SoundFile(
+            str(path), "w", samplerate=SAMPLE_RATE, channels=1, format="wav"
+        ) as f:
+            f.write(audio)
+
+    def _ffmetadata(self, chapters: list[dict[str, Any]]) -> str:
+        lines = [";FFMETADATA1"]
+        for ch in chapters:
+            title = str(ch["title"]).replace("=", "\\=")
+            lines += [
+                "[CHAPTER]",
+                "TIMEBASE=1/1000",
+                f"START={int(ch['start'] * 1000)}",
+                f"END={int(ch['end'] * 1000)}",
+                f"title={title}",
+                "",
+            ]
+        return "\n".join(lines)