Events

class pipecat.services.gemini_multimodal_live.events.MediaChunk(*, mimeType, data)[source]

Bases: BaseModel

Parameters:

mimeType (str)
data (str)

mimeType: str

data: str

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ContentPart(*, text=None, inlineData=None)[source]

Bases: BaseModel

Parameters:

text (str | None)
inlineData (MediaChunk | None)

text: str | None

inlineData: MediaChunk | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.Turn(*, role='user', parts)[source]

Bases: BaseModel

Parameters:

role (Literal['user', 'model'])
parts (List[ContentPart])

role: Literal['user', 'model']

parts: List[ContentPart]

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.StartSensitivity(*values)[source]

Bases: str, Enum

Determines how start of speech is detected.

UNSPECIFIED = 'START_SENSITIVITY_UNSPECIFIED'

HIGH = 'START_SENSITIVITY_HIGH'

LOW = 'START_SENSITIVITY_LOW'

class pipecat.services.gemini_multimodal_live.events.EndSensitivity(*values)[source]

Bases: str, Enum

Determines how end of speech is detected.

UNSPECIFIED = 'END_SENSITIVITY_UNSPECIFIED'

HIGH = 'END_SENSITIVITY_HIGH'

LOW = 'END_SENSITIVITY_LOW'

class pipecat.services.gemini_multimodal_live.events.AutomaticActivityDetection(*, disabled=None, start_of_speech_sensitivity=None, prefix_padding_ms=None, end_of_speech_sensitivity=None, silence_duration_ms=None)[source]

Bases: BaseModel

Configures automatic detection of activity.

Parameters:

disabled (bool | None)
start_of_speech_sensitivity (StartSensitivity | None)
prefix_padding_ms (int | None)
end_of_speech_sensitivity (EndSensitivity | None)
silence_duration_ms (int | None)

disabled: bool | None

start_of_speech_sensitivity: StartSensitivity | None

prefix_padding_ms: int | None

end_of_speech_sensitivity: EndSensitivity | None

silence_duration_ms: int | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.RealtimeInputConfig(*, automatic_activity_detection=None)[source]

Bases: BaseModel

Configures the realtime input behavior.

Parameters:: automatic_activity_detection (AutomaticActivityDetection | None)

automatic_activity_detection: AutomaticActivityDetection | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.RealtimeInput(*, mediaChunks)[source]

Bases: BaseModel

Parameters:: mediaChunks (List[MediaChunk])

mediaChunks: List[MediaChunk]

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ClientContent(*, turns=None, turnComplete=False)[source]

Bases: BaseModel

Parameters:

turns (List[Turn] | None)
turnComplete (bool)

turns: List[Turn] | None

turnComplete: bool

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.AudioInputMessage(*, realtimeInput)[source]

Bases: BaseModel

Parameters:: realtimeInput (RealtimeInput)

realtimeInput: RealtimeInput

classmethod from_raw_audio(raw_audio, sample_rate)[source]

Parameters:

raw_audio (bytes)
sample_rate (int)

Return type:

AudioInputMessage

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.VideoInputMessage(*, realtimeInput)[source]

Bases: BaseModel

Parameters:: realtimeInput (RealtimeInput)

realtimeInput: RealtimeInput

classmethod from_image_frame(frame)[source]

Parameters:: frame (ImageRawFrame)
Return type:: VideoInputMessage

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ClientContentMessage(*, clientContent)[source]

Bases: BaseModel

Parameters:: clientContent (ClientContent)

clientContent: ClientContent

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.SystemInstruction(*, parts)[source]

Bases: BaseModel

Parameters:: parts (List[ContentPart])

parts: List[ContentPart]

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.AudioTranscriptionConfig[source]

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.Setup(*, model, system_instruction=None, tools=None, generation_config=None, input_audio_transcription=None, output_audio_transcription=None, realtime_input_config=None)[source]

Bases: BaseModel

Parameters:

model (str)
system_instruction (SystemInstruction | None)
tools (List[dict] | None)
generation_config (dict | None)
input_audio_transcription (AudioTranscriptionConfig | None)
output_audio_transcription (AudioTranscriptionConfig | None)
realtime_input_config (RealtimeInputConfig | None)

model: str

system_instruction: SystemInstruction | None

tools: List[dict] | None

generation_config: dict | None

input_audio_transcription: AudioTranscriptionConfig | None

output_audio_transcription: AudioTranscriptionConfig | None

realtime_input_config: RealtimeInputConfig | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.Config(*, setup)[source]

Bases: BaseModel

Parameters:: setup (Setup)

setup: Setup

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.SetupComplete[source]

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.InlineData(*, mimeType, data)[source]

Bases: BaseModel

Parameters:

mimeType (str)
data (str)

mimeType: str

data: str

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.Part(*, inlineData=None, text=None)[source]

Bases: BaseModel

Parameters:

inlineData (InlineData | None)
text (str | None)

inlineData: InlineData | None

text: str | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ModelTurn(*, parts)[source]

Bases: BaseModel

Parameters:: parts (List[Part])

parts: List[Part]

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ServerContentInterrupted(*, interrupted)[source]

Bases: BaseModel

Parameters:: interrupted (bool)

interrupted: bool

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ServerContentTurnComplete(*, turnComplete)[source]

Bases: BaseModel

Parameters:: turnComplete (bool)

turnComplete: bool

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.BidiGenerateContentTranscription(*, text)[source]

Bases: BaseModel

Parameters:: text (str)

text: str

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ServerContent(*, modelTurn=None, interrupted=None, turnComplete=None, inputTranscription=None, outputTranscription=None)[source]

Bases: BaseModel

Parameters:

modelTurn (ModelTurn | None)
interrupted (bool | None)
turnComplete (bool | None)
inputTranscription (BidiGenerateContentTranscription | None)
outputTranscription (BidiGenerateContentTranscription | None)

modelTurn: ModelTurn | None

interrupted: bool | None

turnComplete: bool | None

inputTranscription: BidiGenerateContentTranscription | None

outputTranscription: BidiGenerateContentTranscription | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.FunctionCall(*, id, name, args)[source]

Bases: BaseModel

Parameters:

id (str)
name (str)
args (dict)

id: str

name: str

args: dict

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ToolCall(*, functionCalls)[source]

Bases: BaseModel

Parameters:: functionCalls (List[FunctionCall])

functionCalls: List[FunctionCall]

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.Modality(*values)[source]

Bases: str, Enum

Modality types in token counts.

UNSPECIFIED = 'MODALITY_UNSPECIFIED'

TEXT = 'TEXT'

IMAGE = 'IMAGE'

AUDIO = 'AUDIO'

VIDEO = 'VIDEO'

class pipecat.services.gemini_multimodal_live.events.ModalityTokenCount(*, modality, tokenCount)[source]

Bases: BaseModel

Token count for a specific modality.

Parameters:

modality (Modality)
tokenCount (int)

modality: Modality

tokenCount: int

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.UsageMetadata(*, promptTokenCount=None, cachedContentTokenCount=None, responseTokenCount=None, toolUsePromptTokenCount=None, thoughtsTokenCount=None, totalTokenCount=None, promptTokensDetails=None, cacheTokensDetails=None, responseTokensDetails=None, toolUsePromptTokensDetails=None)[source]

Bases: BaseModel

Usage metadata about the response.

Parameters:

promptTokenCount (int | None)
cachedContentTokenCount (int | None)
responseTokenCount (int | None)
toolUsePromptTokenCount (int | None)
thoughtsTokenCount (int | None)
totalTokenCount (int | None)
promptTokensDetails (List[ModalityTokenCount] | None)
cacheTokensDetails (List[ModalityTokenCount] | None)
responseTokensDetails (List[ModalityTokenCount] | None)
toolUsePromptTokensDetails (List[ModalityTokenCount] | None)

promptTokenCount: int | None

cachedContentTokenCount: int | None

responseTokenCount: int | None

toolUsePromptTokenCount: int | None

thoughtsTokenCount: int | None

totalTokenCount: int | None

promptTokensDetails: List[ModalityTokenCount] | None

cacheTokensDetails: List[ModalityTokenCount] | None

responseTokensDetails: List[ModalityTokenCount] | None

toolUsePromptTokensDetails: List[ModalityTokenCount] | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class pipecat.services.gemini_multimodal_live.events.ServerEvent(*, setupComplete=None, serverContent=None, toolCall=None, usageMetadata=None)[source]

Bases: BaseModel

Parameters:

setupComplete (SetupComplete | None)
serverContent (ServerContent | None)
toolCall (ToolCall | None)
usageMetadata (UsageMetadata | None)

setupComplete: SetupComplete | None

serverContent: ServerContent | None

toolCall: ToolCall | None

usageMetadata: UsageMetadata | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

pipecat.services.gemini_multimodal_live.events.parse_server_event(str)[source]

class pipecat.services.gemini_multimodal_live.events.ContextWindowCompressionConfig(*, sliding_window=True, trigger_tokens=None)[source]

Bases: BaseModel

Configuration for context window compression.

Parameters:

sliding_window (bool | None)
trigger_tokens (int | None)

sliding_window: bool | None

trigger_tokens: int | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].