diff --git a/run.py b/run.py index ac5cbbcc3..0b6f61433 100644 --- a/run.py +++ b/run.py @@ -1,6 +1,5 @@ import argparse import asyncio -import base64 import json import multiprocessing import os @@ -13,7 +12,7 @@ from io import BytesIO, TextIOWrapper from pathlib import Path from tempfile import NamedTemporaryFile, TemporaryFile -from typing import Annotated, Any, Literal, Optional +from typing import Annotated, Any, Optional import soundfile import uvicorn @@ -24,7 +23,6 @@ from fastapi.openapi.utils import get_openapi from fastapi.responses import JSONResponse from fastapi.templating import Jinja2Templates -from pydantic import parse_obj_as from starlette.background import BackgroundTask from starlette.middleware.errors import ServerErrorMiddleware from starlette.responses import FileResponse @@ -34,7 +32,7 @@ check_disabled_mutable_api, deprecated_mutable_api, ) -from voicevox_engine.app.routers import preset, setting, user_dict +from voicevox_engine.app.routers import preset, setting, speaker, user_dict from voicevox_engine.cancellable_engine import CancellableEngine from voicevox_engine.core.core_adapter import CoreAdapter from voicevox_engine.core.core_initializer import initialize_cores @@ -42,11 +40,7 @@ from voicevox_engine.engine_manifest.EngineManifestLoader import EngineManifestLoader from voicevox_engine.library_manager import LibraryManager from voicevox_engine.metas.Metas import StyleId -from voicevox_engine.metas.MetasStore import ( - MetasStore, - construct_lookup, - filter_speakers_and_styles, -) +from voicevox_engine.metas.MetasStore import MetasStore, construct_lookup from voicevox_engine.model import ( AccentPhrase, AudioQuery, @@ -58,8 +52,6 @@ ParseKanaBadRequest, ParseKanaError, Score, - Speaker, - SpeakerInfo, StyleIdNotFoundError, SupportedDevicesInfo, VvlibManifest, @@ -91,10 +83,6 @@ from voicevox_engine.utility.run_utility import decide_boolean_from_env -def b64encode_str(s: bytes) -> str: - return base64.b64encode(s).decode("utf-8") - - def set_output_log_utf8() -> None: """ stdout/stderrのエンコーディングをUTF-8に切り替える関数 @@ -755,144 +743,7 @@ async def core_versions() -> Response: media_type="application/json", ) - @app.get("/speakers", response_model=list[Speaker], tags=["その他"]) - def speakers( - core_version: str | None = None, - ) -> list[Speaker]: - speakers = metas_store.load_combined_metas(get_core(core_version)) - return filter_speakers_and_styles(speakers, "speaker") - - @app.get("/speaker_info", response_model=SpeakerInfo, tags=["その他"]) - def speaker_info( - speaker_uuid: str, - core_version: str | None = None, - ) -> SpeakerInfo: - """ - 指定されたspeaker_uuidに関する情報をjson形式で返します。 - 画像や音声はbase64エンコードされたものが返されます。 - """ - return _speaker_info( - speaker_uuid=speaker_uuid, - speaker_or_singer="speaker", - core_version=core_version, - ) - - # FIXME: この関数をどこかに切り出す - def _speaker_info( - speaker_uuid: str, - speaker_or_singer: Literal["speaker", "singer"], - core_version: str | None, - ) -> SpeakerInfo: - # エンジンに含まれる話者メタ情報は、次のディレクトリ構造に従わなければならない： - # {root_dir}/ - # speaker_info/ - # {speaker_uuid_0}/ - # policy.md - # portrait.png - # icons/ - # {id_0}.png - # {id_1}.png - # ... - # portraits/ - # {id_0}.png - # {id_1}.png - # ... - # voice_samples/ - # {id_0}_001.wav - # {id_0}_002.wav - # {id_0}_003.wav - # {id_1}_001.wav - # ... - # {speaker_uuid_1}/ - # ... - - # 該当話者の検索 - speakers = parse_obj_as( - list[Speaker], json.loads(get_core(core_version).speakers) - ) - speakers = filter_speakers_and_styles(speakers, speaker_or_singer) - for i in range(len(speakers)): - if speakers[i].speaker_uuid == speaker_uuid: - speaker = speakers[i] - break - else: - raise HTTPException(status_code=404, detail="該当する話者が見つかりません") - - try: - speaker_path = root_dir / "speaker_info" / speaker_uuid - # 話者情報の取得 - # speaker policy - policy_path = speaker_path / "policy.md" - policy = policy_path.read_text("utf-8") - # speaker portrait - portrait_path = speaker_path / "portrait.png" - portrait = b64encode_str(portrait_path.read_bytes()) - # スタイル情報の取得 - style_infos = [] - for style in speaker.styles: - id = style.id - # style icon - style_icon_path = speaker_path / "icons" / f"{id}.png" - icon = b64encode_str(style_icon_path.read_bytes()) - # style portrait - style_portrait_path = speaker_path / "portraits" / f"{id}.png" - style_portrait = None - if style_portrait_path.exists(): - style_portrait = b64encode_str(style_portrait_path.read_bytes()) - # voice samples - voice_samples = [ - b64encode_str( - ( - speaker_path - / "voice_samples/{}_{}.wav".format(id, str(j + 1).zfill(3)) - ).read_bytes() - ) - for j in range(3) - ] - style_infos.append( - { - "id": id, - "icon": icon, - "portrait": style_portrait, - "voice_samples": voice_samples, - } - ) - except FileNotFoundError: - import traceback - - traceback.print_exc() - raise HTTPException( - status_code=500, detail="追加情報が見つかりませんでした" - ) - - ret_data = SpeakerInfo( - policy=policy, - portrait=portrait, - style_infos=style_infos, - ) - return ret_data - - @app.get("/singers", response_model=list[Speaker], tags=["その他"]) - def singers( - core_version: str | None = None, - ) -> list[Speaker]: - singers = metas_store.load_combined_metas(get_core(core_version)) - return filter_speakers_and_styles(singers, "singer") - - @app.get("/singer_info", response_model=SpeakerInfo, tags=["その他"]) - def singer_info( - speaker_uuid: str, - core_version: str | None = None, - ) -> SpeakerInfo: - """ - 指定されたspeaker_uuidに関する情報をjson形式で返します。 - 画像や音声はbase64エンコードされたものが返されます。 - """ - return _speaker_info( - speaker_uuid=speaker_uuid, - speaker_or_singer="singer", - core_version=core_version, - ) + app.include_router(speaker.generate_router(get_core, metas_store, root_dir)) if engine_manifest_data.supported_features.manage_library: diff --git a/voicevox_engine/app/routers/speaker.py b/voicevox_engine/app/routers/speaker.py new file mode 100644 index 000000000..1e539224b --- /dev/null +++ b/voicevox_engine/app/routers/speaker.py @@ -0,0 +1,167 @@ +"""話者情報機能を提供する API Router""" + +import base64 +import json +from pathlib import Path +from typing import Callable, Literal + +from fastapi import APIRouter, HTTPException +from pydantic import parse_obj_as + +from voicevox_engine.core.core_adapter import CoreAdapter +from voicevox_engine.metas.MetasStore import MetasStore, filter_speakers_and_styles +from voicevox_engine.model import Speaker, SpeakerInfo + + +def b64encode_str(s: bytes) -> str: + return base64.b64encode(s).decode("utf-8") + + +def generate_router( + get_core: Callable[[str | None], CoreAdapter], + metas_store: MetasStore, + root_dir: Path, +) -> APIRouter: + """話者情報 API Router を生成する""" + router = APIRouter() + + @router.get("/speakers", response_model=list[Speaker], tags=["その他"]) + def speakers( + core_version: str | None = None, + ) -> list[Speaker]: + speakers = metas_store.load_combined_metas(get_core(core_version)) + return filter_speakers_and_styles(speakers, "speaker") + + @router.get("/speaker_info", response_model=SpeakerInfo, tags=["その他"]) + def speaker_info( + speaker_uuid: str, + core_version: str | None = None, + ) -> SpeakerInfo: + """ + 指定されたspeaker_uuidに関する情報をjson形式で返します。 + 画像や音声はbase64エンコードされたものが返されます。 + """ + return _speaker_info( + speaker_uuid=speaker_uuid, + speaker_or_singer="speaker", + core_version=core_version, + ) + + # FIXME: この関数をどこかに切り出す + def _speaker_info( + speaker_uuid: str, + speaker_or_singer: Literal["speaker", "singer"], + core_version: str | None, + ) -> SpeakerInfo: + # エンジンに含まれる話者メタ情報は、次のディレクトリ構造に従わなければならない： + # {root_dir}/ + # speaker_info/ + # {speaker_uuid_0}/ + # policy.md + # portrait.png + # icons/ + # {id_0}.png + # {id_1}.png + # ... + # portraits/ + # {id_0}.png + # {id_1}.png + # ... + # voice_samples/ + # {id_0}_001.wav + # {id_0}_002.wav + # {id_0}_003.wav + # {id_1}_001.wav + # ... + # {speaker_uuid_1}/ + # ... + + # 該当話者の検索 + speakers = parse_obj_as( + list[Speaker], json.loads(get_core(core_version).speakers) + ) + speakers = filter_speakers_and_styles(speakers, speaker_or_singer) + for i in range(len(speakers)): + if speakers[i].speaker_uuid == speaker_uuid: + speaker = speakers[i] + break + else: + raise HTTPException(status_code=404, detail="該当する話者が見つかりません") + + try: + speaker_path = root_dir / "speaker_info" / speaker_uuid + # 話者情報の取得 + # speaker policy + policy_path = speaker_path / "policy.md" + policy = policy_path.read_text("utf-8") + # speaker portrait + portrait_path = speaker_path / "portrait.png" + portrait = b64encode_str(portrait_path.read_bytes()) + # スタイル情報の取得 + style_infos = [] + for style in speaker.styles: + id = style.id + # style icon + style_icon_path = speaker_path / "icons" / f"{id}.png" + icon = b64encode_str(style_icon_path.read_bytes()) + # style portrait + style_portrait_path = speaker_path / "portraits" / f"{id}.png" + style_portrait = None + if style_portrait_path.exists(): + style_portrait = b64encode_str(style_portrait_path.read_bytes()) + # voice samples + voice_samples = [ + b64encode_str( + ( + speaker_path + / "voice_samples/{}_{}.wav".format(id, str(j + 1).zfill(3)) + ).read_bytes() + ) + for j in range(3) + ] + style_infos.append( + { + "id": id, + "icon": icon, + "portrait": style_portrait, + "voice_samples": voice_samples, + } + ) + except FileNotFoundError: + import traceback + + traceback.print_exc() + raise HTTPException( + status_code=500, detail="追加情報が見つかりませんでした" + ) + + ret_data = SpeakerInfo( + policy=policy, + portrait=portrait, + style_infos=style_infos, + ) + return ret_data + + @router.get("/singers", response_model=list[Speaker], tags=["その他"]) + def singers( + core_version: str | None = None, + ) -> list[Speaker]: + singers = metas_store.load_combined_metas(get_core(core_version)) + return filter_speakers_and_styles(singers, "singer") + + @router.get("/singer_info", response_model=SpeakerInfo, tags=["その他"]) + def singer_info( + speaker_uuid: str, + core_version: str | None = None, + ) -> SpeakerInfo: + """ + 指定されたspeaker_uuidに関する情報をjson形式で返します。 + 画像や音声はbase64エンコードされたものが返されます。 + """ + return _speaker_info( + speaker_uuid=speaker_uuid, + speaker_or_singer="singer", + core_version=core_version, + ) + + return router