Pipecat C++ Client SDK 1.3.0
Connect native apps to Pipecat bots
Loading...
Searching...
No Matches
pipecat::rtvi::UserInputData Struct Reference

What the user said or typed. More...

#include <pipecat/rtvi/messages.h>

Public Attributes

std::string text
 The words.
 
std::string input_type
 
bool final = false
 
std::string timestamp
 When it was said or sent, as an ISO 8601 date.
 
std::optional< std::string > user_id
 
std::optional< std::string > msg_id
 For chat, the ID PipecatClient::send_text() returned.
 
std::vector< UserPerceptionSignal > perception
 
std::optional< TranscriptionSpeaker > speaker
 

Detailed Description

What the user said or typed.

Member Data Documentation

◆ final

bool pipecat::rtvi::UserInputData::final = false

Whether the text is final. Only a transcription can be not final, and a new one will replace it.

◆ input_type

std::string pipecat::rtvi::UserInputData::input_type

What kind of input it is: "transcription" for what the user said, "chat" for text sent with PipecatClient::send_text(), or "backchannel" for a short acknowledgment the user said while the bot talked, like "mhm", which doesn't interrupt the bot.

◆ perception

std::vector<UserPerceptionSignal> pipecat::rtvi::UserInputData::perception

For a transcription, how the user said it, e.g. their tone, if the bot reads it. Interim transcriptions can have it too. Only from bots with expressive conversation (alpha).

◆ speaker

std::optional<TranscriptionSpeaker> pipecat::rtvi::UserInputData::speaker

For a transcription, who said it, if the bot's speech-to-text tells speakers apart. Unlike user_id, it tells apart people sharing one microphone.

◆ user_id

std::optional<std::string> pipecat::rtvi::UserInputData::user_id

ID of the user who spoke or typed. For chat, only with transports that know who sent it, like Daily.


The documentation for this struct was generated from the following file: