audex 1.0.7a3__py3-none-any.whl

audex/config/helper/provider/dashscope.py

@@ -0,0 +1,13 @@
+from __future__ import annotations
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+from audex.utils import UNSET
+from audex.utils import Unset
+
+
+class DashscopeCredentialConfig(BaseModel):
+    api_key: str | Unset = Field(
+        default=UNSET,
+        description="Dashscope API key for authentication.",
+    )

audex/config/helper/provider/unisound.py

@@ -0,0 +1,18 @@
+from __future__ import annotations
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+from audex.utils import UNSET
+from audex.utils import Unset
+
+
+class UnisoundCredentialConfig(BaseModel):
+    appkey: str | Unset = Field(
+        default=UNSET,
+        description="Unisound App Key",
+    )
+
+    secret: str | Unset = Field(
+        default=UNSET,
+        description="Unisound Secret",
+    )

audex/config/helper/provider/xfyun.py

@@ -0,0 +1,23 @@
+from __future__ import annotations
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+from audex.utils import UNSET
+from audex.utils import Unset
+
+
+class XFYunCredentialConfig(BaseModel):
+    app_id: str | Unset = Field(
+        default=UNSET,
+        description="XFYun Application ID. Must be provided for authentication.",
+    )
+
+    api_key: str | Unset = Field(
+        default=UNSET,
+        description="XFYun API Key. Must be provided for authentication.",
+    )
+
+    api_secret: str | Unset = Field(
+        default=UNSET,
+        description="XFYun API Secret. Must be provided for authentication.",
+    )

audex/config/infrastructure/__init__.py

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+from pydantic import Field
+
+from audex.config.infrastructure.cache import CacheConfig
+from audex.config.infrastructure.database import SQLiteConfig
+from audex.config.infrastructure.recorder import RecorderConfig
+from audex.config.infrastructure.store import StoreConfig
+from audex.helper.settings import BaseModel
+
+
+class InfrastructureConfig(BaseModel):
+    sqlite: SQLiteConfig = Field(
+        default_factory=SQLiteConfig,
+        description="SQLite configuration.",
+    )
+
+    cache: CacheConfig = Field(
+        default_factory=CacheConfig,
+        description="Cache configuration.",
+    )
+
+    store: StoreConfig = Field(
+        default_factory=StoreConfig,
+        description="Store configuration.",
+    )
+
+    recorder: RecorderConfig = Field(
+        default_factory=RecorderConfig,
+        description="Recorder configuration.",
+    )

audex/config/infrastructure/cache.py

@@ -0,0 +1,51 @@
+from __future__ import annotations
+
+import typing as t
+
+from audex import __prog__
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class InmemoryCacheConfig(BaseModel):
+    cache_type: t.Literal["lru", "lfu", "ttl", "fifo"] = Field(
+        default="ttl",
+        description="Type of in-memory cache algorithm to use.",
+    )
+
+    max_size: int = Field(
+        default=1024,
+        description="Maximum number of items to store in the cache.",
+    )
+
+    default_ttl: int = Field(
+        default=300,
+        description="Default time-to-live (TTL) for cache items in seconds.",
+    )
+
+    negative_ttl: int = Field(
+        default=60,
+        description="TTL for negative cache entries in seconds.",
+    )
+
+
+class CacheConfig(BaseModel):
+    provider: t.Literal["inmemory"] = Field(
+        default="inmemory",
+        description="Type of cache backend to use.",
+    )
+
+    split_char: str = Field(
+        default=":",
+        description="Character used to split cache keys.",
+    )
+
+    prefix: str = Field(
+        default=__prog__,
+        description="Prefix for all cache keys.",
+    )
+
+    inmemory: InmemoryCacheConfig = Field(
+        default_factory=InmemoryCacheConfig,
+        description="Configuration for in-memory cache backend.",
+    )

audex/config/infrastructure/database.py

@@ -0,0 +1,48 @@
+from __future__ import annotations
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class SQLiteConfig(BaseModel):
+    uri: str = Field(
+        default="sqlite+aiosqlite:///./audex.db",
+        description="SQLite database URI.",
+        windows_default="sqlite+aiosqlite:///C:/ProgramData/Audex/audex.db",
+        linux_default="sqlite+aiosqlite://${HOME}/.local/share/audex/audex.db",
+    )
+
+    echo: bool = Field(
+        default=False,
+        description="Enable SQL query logging for debugging purposes.",
+    )
+
+    pool_size: int = Field(
+        default=20,
+        description="The size of the database connection pool.",
+    )
+
+    max_overflow: int = Field(
+        default=10,
+        description="The maximum overflow size of the connection pool.",
+    )
+
+    pool_timeout: float = Field(
+        default=30.0,
+        description="The timeout in seconds to wait for a connection from the pool.",
+    )
+
+    pool_recycle: int = Field(
+        default=3600,
+        description="The number of seconds after which a connection is automatically recycled.",
+    )
+
+    pool_pre_ping: bool = Field(
+        default=True,
+        description="Enable connection health checks before using a connection from the pool.",
+    )
+
+    create_all: bool = Field(
+        default=True,
+        description="Whether to create all tables on initialization.",
+    )

audex/config/infrastructure/recorder.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+import typing as t
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class RecorderConfig(BaseModel):
+    format: t.Literal["float32", "int32", "int16", "int8", "uint8"] = Field(
+        default="int16",
+        description="Audio sample format.",
+    )
+
+    channels: int = Field(
+        default=1,
+        ge=1,
+        le=8,
+        description="Number of audio channels.",
+    )
+
+    chunk: int = Field(
+        default=1024,
+        ge=256,
+        le=8192,
+        description="Number of frames per buffer.",
+    )
+
+    input_device_index: int | None = Field(
+        default=None,
+        description="Index of the input audio device. If None, the default device is used.",
+    )

audex/config/infrastructure/store.py

@@ -0,0 +1,19 @@
+from __future__ import annotations
+
+import typing as t
+
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class StoreConfig(BaseModel):
+    type: t.Literal["localfile"] = Field(
+        default="localfile",
+        description="Type of the store.",
+    )
+
+    base_url: str = Field(
+        default="store",
+        description="Store base URL. In case of 'file' type, it is the local directory path. In case of 'obs' type, it is the OBS bucket URL.",
+        system_path_type="data",
+    )

audex/config/provider/__init__.py

@@ -0,0 +1,18 @@
+from __future__ import annotations
+
+from audex.config.provider.transcription import TranscriptionConfig
+from audex.config.provider.vpr import VPRConfig
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class ProviderConfig(BaseModel):
+    transcription: TranscriptionConfig = Field(
+        default_factory=TranscriptionConfig,
+        description="Transcription provider configuration",
+    )
+
+    vpr: VPRConfig = Field(
+        default_factory=VPRConfig,
+        description="VPR provider configuration",
+    )

audex/config/provider/transcription.py

@@ -0,0 +1,109 @@
+from __future__ import annotations
+
+import typing as t
+
+from audex.config.helper.client.websocket import WebsocketClientConfig
+from audex.config.helper.provider.dashscope import DashscopeCredentialConfig
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class DashscopeTranscriptionSessionConfig(BaseModel):
+    fmt: t.Literal["pcm", "wav", "mp3", "opus", "speex", "aac", "amr"] = Field(
+        default="pcm",
+        description="Audio format for transcription session.",
+    )
+
+    silence_duration_ms: int | None = Field(
+        default=None,
+        description="Duration of silence in milliseconds to consider as end of speech.",
+    )
+
+    vocabulary_id: str | None = Field(
+        default=None,
+        description="Custom vocabulary ID for transcription.",
+    )
+
+    disfluency_removal_enabled: bool | None = Field(
+        default=None,
+        description="Enable disfluency removal in transcription.",
+    )
+
+    lang_hints: list[t.Literal["zh", "en", "ja", "yue", "ko", "de", "fr", "ru"]] | None = Field(
+        default=None,
+        description="Language hints for transcription.",
+    )
+
+    semantic_punctuation: bool | None = Field(
+        default=None,
+        description="Enable semantic punctuation in transcription.",
+    )
+
+    multi_thres_mode: bool | None = Field(
+        default=None,
+        description="Enable multi-threshold mode for transcription.",
+    )
+
+    punctuation_pred: bool | None = Field(
+        default=None,
+        description="Enable punctuation prediction in transcription.",
+    )
+
+    heartbeat: bool | None = Field(
+        default=None,
+        description="Enable heartbeat messages during transcription.",
+    )
+
+    itn: bool | None = Field(
+        default=None,
+        description="Enable inverse text normalization in transcription.",
+    )
+
+    resources: list[str] | None = Field(
+        default=None,
+        description="List of resource identifiers for transcription.",
+    )
+
+
+class DashscopeTranscriptionConfig(BaseModel):
+    credential: DashscopeCredentialConfig = Field(
+        default_factory=DashscopeCredentialConfig,
+        description="Credentials for Dashscope API access.",
+    )
+
+    model: str = Field(
+        default="paraformer-realtime-v2",
+        description="Dashscope transcription model to use.",
+    )
+
+    user_agent: str | None = Field(
+        default=None,
+        description="Custom User-Agent header for API requests.",
+    )
+
+    workspace: str | None = Field(
+        default=None,
+        description="Workspace identifier for Dashscope service.",
+    )
+
+    websocket: WebsocketClientConfig = Field(
+        default_factory=WebsocketClientConfig,
+        description="WebSocket client configuration for Dashscope.",
+    )
+
+    session: DashscopeTranscriptionSessionConfig = Field(
+        default_factory=DashscopeTranscriptionSessionConfig,
+        description="Session configuration for Dashscope transcription.",
+    )
+
+
+class TranscriptionConfig(BaseModel):
+    provider: t.Literal["dashscope"] = Field(
+        default="dashscope",
+        description="The transcription service provider.",
+    )
+
+    dashscope: DashscopeTranscriptionConfig = Field(
+        default_factory=DashscopeTranscriptionConfig,
+        description="Configuration for Dashscope transcription service.",
+    )

audex/config/provider/vpr.py

@@ -0,0 +1,99 @@
+from __future__ import annotations
+
+import contextlib
+import os
+import pathlib
+import typing as t
+
+from pydantic import model_validator
+
+from audex.config.helper.client.http import HttpClientConfig
+from audex.config.helper.provider.unisound import UnisoundCredentialConfig
+from audex.config.helper.provider.xfyun import XFYunCredentialConfig
+from audex.helper.settings import BaseModel
+from audex.helper.settings.fields import Field
+
+
+class XFYunVPRConfig(BaseModel):
+    credential: XFYunCredentialConfig = Field(
+        default_factory=XFYunCredentialConfig,
+        description="Credentials for XFYun VPR service.",
+    )
+
+    group_id: str | None = Field(
+        default=None,
+        description="The group ID for voiceprint recognition.",
+    )
+
+    group_id_path: os.PathLike[str] = Field(
+        default=pathlib.Path(".xf.vprgroup"),
+        description="The file path to read the group ID from if not provided directly.",
+        system_default=".xf.vprgroup",
+        system_path_type="config",
+    )
+
+    http: HttpClientConfig = Field(
+        default_factory=HttpClientConfig,
+        description="HTTP client configuration for XFYun VPR service.",
+    )
+
+    @model_validator(mode="after")
+    def load_group_id_from_file(self) -> t.Self:
+        if self.group_id is None:
+            with (
+                contextlib.suppress(FileNotFoundError),
+                pathlib.Path(self.group_id_path).open("r") as f,
+            ):
+                self.group_id = f.read().strip()
+        return self
+
+
+class UnisoundVPRConfig(BaseModel):
+    credential: UnisoundCredentialConfig = Field(
+        default_factory=UnisoundCredentialConfig,
+        description="Credentials for Unisound VPR service.",
+    )
+
+    group_id: str | None = Field(
+        default=None,
+        description="The group ID for voiceprint recognition.",
+    )
+
+    group_id_path: os.PathLike[str] = Field(
+        default=pathlib.Path(".unisound.vprgroup"),
+        description="The file path to read the group ID from if not provided directly.",
+        system_default=".unisound.vprgroup",
+        system_path_type="config",
+    )
+
+    http: HttpClientConfig = Field(
+        default_factory=HttpClientConfig,
+        description="HTTP client configuration for Unisound VPR service.",
+    )
+
+    @model_validator(mode="after")
+    def load_group_id_from_file(self) -> t.Self:
+        if self.group_id is None:
+            with (
+                contextlib.suppress(FileNotFoundError),
+                pathlib.Path(self.group_id_path).open("r") as f,
+            ):
+                self.group_id = f.read().strip()
+        return self
+
+
+class VPRConfig(BaseModel):
+    provider: t.Literal["xfyun", "unisound"] = Field(
+        default="xfyun",
+        description="The VPR service provider to use.",
+    )
+
+    xfyun: XFYunVPRConfig = Field(
+        default_factory=XFYunVPRConfig,
+        description="Configuration for the XFYun VPR service.",
+    )
+
+    unisound: UnisoundVPRConfig = Field(
+        default_factory=UnisoundVPRConfig,
+        description="Configuration for the Unisound VPR service.",
+    )

audex/container.py

@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+from dependency_injector import containers
+from dependency_injector import providers
+
+from audex.injectors.config import config
+from audex.injectors.lifespan import lifespan
+from audex.lib.injectors.container import InfrastructureContainer
+from audex.service.injectors.container import ServiceContainer
+from audex.view.container import ViewContainer
+
+
+class Container(containers.DeclarativeContainer):
+    # Configuration
+    config = providers.Callable(config)
+
+    # Containers
+    infrastructure = providers.Container(InfrastructureContainer, config=config)
+    service = providers.Container(
+        ServiceContainer,
+        config=config,
+        infrastructure=infrastructure,
+        repository=infrastructure.repository,
+    )
+
+    # Lifespan
+    lifespan = providers.Singleton(
+        lifespan,
+        config,
+        infrastructure.session_manager,
+        infrastructure.wifi,
+        infrastructure.cache,
+        infrastructure.sqlite,
+        infrastructure.vpr,
+        infrastructure.recorder,
+        infrastructure.transcription,
+    )
+
+    # Views
+    views = providers.Container(ViewContainer, config=config, service=service, lifespan=lifespan)