Constants

gllm_multimodal.constants

Constants for multimodal operations in Gen AI applications.

DEFAULT_ROUTER_EMBEDDING_MODEL = 'twelvelabs/marengo3.0' module-attribute

AudioFormatConstants

Constants for audio format related keys and MIME types.

FORMAT_FLAC = 'flac' class-attribute instance-attribute

FORMAT_M4A = 'm4a' class-attribute instance-attribute

FORMAT_MP3 = 'mp3' class-attribute instance-attribute

FORMAT_MPEG = 'mpeg' class-attribute instance-attribute

FORMAT_MPGA = 'mpga' class-attribute instance-attribute

FORMAT_OGA = 'oga' class-attribute instance-attribute

FORMAT_OGG = 'ogg' class-attribute instance-attribute

FORMAT_TO_MIME_TYPE = {FORMAT_MP3: MIME_MPEG, FORMAT_MPEG: MIME_MPEG, FORMAT_MPGA: MIME_MPEG, FORMAT_M4A: MIME_MP4, FORMAT_WAV: MIME_WAV, FORMAT_WAVE: MIME_WAV, FORMAT_FLAC: MIME_FLAC, FORMAT_OGG: MIME_OGG, FORMAT_OGA: MIME_OGG} class-attribute instance-attribute

FORMAT_WAV = 'wav' class-attribute instance-attribute

FORMAT_WAVE = 'wave' class-attribute instance-attribute

MIME_FLAC = 'audio/flac' class-attribute instance-attribute

MIME_MP4 = 'audio/mp4' class-attribute instance-attribute

MIME_MPEG = 'audio/mpeg' class-attribute instance-attribute

MIME_OGG = 'audio/ogg' class-attribute instance-attribute

MIME_WAV = 'audio/wav' class-attribute instance-attribute

CaptionConstants

Constants for caption operations in Gen AI applications.

ATTACHMENTS_CONTEXT = 'attachments_context' class-attribute instance-attribute

CAPTION_DEFAULT_JSON_KEY = 'captions' class-attribute instance-attribute

DEFAULT_NUMBER_OF_CAPTIONS = 5 class-attribute instance-attribute

DOMAIN_KNOWLEDGE = 'domain_knowledge' class-attribute instance-attribute

IMAGE_DESCRIPTION = 'image_description' class-attribute instance-attribute

IMAGE_METADATA = 'image_metadata' class-attribute instance-attribute

IMAGE_ONE_LINER = 'image_one_liner' class-attribute instance-attribute

IS_STRUCTURED_CAPTION = 'is_structured_caption' class-attribute instance-attribute

NUMBER_OF_CAPTIONS = 'number_of_captions' class-attribute instance-attribute

CaptionOutputFormattingStrategy

Bases: StrEnum

Defines how output should be formatted.

STRUCTURED = 'structured' class-attribute instance-attribute

TEXT = 'text' class-attribute instance-attribute

ConverterResultTag

Bases: StrEnum

Defines converter result tags.

CAPTION = 'captions' class-attribute instance-attribute

MERMAID = 'mermaid' class-attribute instance-attribute

STRUCTURED_CAPTION = 'structured_caption' class-attribute instance-attribute

ExifConstants

Constants for EXIF tag operations in image metadata extraction.

GPS_LATITUDE = 'GPS GPSLatitude' class-attribute instance-attribute

GPS_LATITUDE_REF = 'GPS GPSLatitudeRef' class-attribute instance-attribute

GPS_LONGITUDE = 'GPS GPSLongitude' class-attribute instance-attribute

GPS_LONGITUDE_REF = 'GPS GPSLongitudeRef' class-attribute instance-attribute

GPS_SOUTH = 'S' class-attribute instance-attribute

GPS_WEST = 'W' class-attribute instance-attribute

LATITUDE = 'latitude' class-attribute instance-attribute

LONGITUDE = 'longitude' class-attribute instance-attribute

ImageToTextConstants

Constants for image-to-text operations in Gen AI applications.

FILENAME = 'filename' class-attribute instance-attribute

NOT_GIVEN = 'not_given' class-attribute instance-attribute

NO_IMAGE_FILENAME = 'no_image_filename' class-attribute instance-attribute

Modality

Bases: StrEnum

Defines supported modalities.

AUDIO = 'audio' class-attribute instance-attribute

IMAGE = 'image' class-attribute instance-attribute

TEXT = 'text' class-attribute instance-attribute

VIDEO = 'video' class-attribute instance-attribute

ModalityConverterApproach

Bases: StrEnum

Defines supported modality converter approach types.

GEMINI = 'gemini' class-attribute instance-attribute

GOOGLE_CLOUD = 'google_cloud' class-attribute instance-attribute

LM_BASED = 'lm_based' class-attribute instance-attribute

PROSA = 'prosa' class-attribute instance-attribute

WHISPER = 'whisper' class-attribute instance-attribute

YOUTUBE = 'youtube' class-attribute instance-attribute

ModalityConverterBuildStrategy

Bases: StrEnum

Defines how a modality converter should be constructed from its configuration.

KWARGS = 'kwargs' class-attribute instance-attribute

LMRP = 'lmrp' class-attribute instance-attribute

PRESET = 'preset' class-attribute instance-attribute

ModalityConverterTask

Bases: StrEnum

Defines supported modality converter tasks.

AUTO = 'auto' class-attribute instance-attribute

CAPTIONING = 'captioning' class-attribute instance-attribute

MERMAID = 'mermaid' class-attribute instance-attribute

TRANSCRIPT = 'transcript' class-attribute instance-attribute

ModalityTransformerRouterPreset

Bases: StrEnum

Defines supported modality transformer router presets.

DOMAIN_SPECIFIC = 'domain_specific' class-attribute instance-attribute

MULTIMODAL = 'multimodal' class-attribute instance-attribute

ModalityTransformerType

Bases: StrEnum

Defines supported modality transformer types.

GENERIC = 'generic' class-attribute instance-attribute

STANDARD = 'standard' class-attribute instance-attribute

WhisperResponseKeys

String keys used when parsing OpenAI Whisper API responses.

AUDIO_TRANSCRIPTS = 'audio_transcripts' class-attribute instance-attribute

LANGUAGE = 'language' class-attribute instance-attribute

SEGMENTS = 'segments' class-attribute instance-attribute

TEXT = 'text' class-attribute instance-attribute

WORDS = 'words' class-attribute instance-attribute