Protocol Documentation

unique id to identify the client connection

response message

API response status code as defined in `APIStatus`

Complete ASR Response in Riva Skills ASR result schema

start time in ISO8601 format, e.g. 2024-03-08T13:33:30.736Z

stop time in ISO8601 format

unique id to identify the client connection

chat engine result

unique id to identify the client connection

bot name with version like {bot_name}_v{bot_version}, e.g chitchat_bot_v1.

query

unique id for identifying the query

user id

The language of the supplied query string as a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag. Example: "en-US".

The language of the response required from chat engine as a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag. Example: "en-US".

Flag to send standalone text requests, when set true reponse is not sent to TTS when set to false reponse will be sent to TTS

key-value pair for user context to be sent to chat engine

key-value pair for meta data to be sent to chat engine

unique id to identify the client connection

query

unique id for identifying the query

user id

session id if generated by chat engine

chat engine response for the query passed in `ChatRequest`

chat engine cleaned up response text after markdown language tags removal

flag to indicate whether this is final response or intermediate response, when true there will be no more responses for the requested `ChatRequest`

chat engine response in json format

bot name with version like {bot_name}_v{bot_version}, e.g chitchat_bot_v1.

unique id to identify the client connection

bot name with version like {bot_name}_v{bot_version}, e.g chitchat_bot_v1.

event type

unique event id

user id

key-value pair for user context to be sent to chat engine

key-value pair for meta data to be sent to chat engine

unique id to identify the client connection

event type

unique event id

user id

text response

unique id to identify the client connection

unique id to identify the client connection

A unique id sent by the client to identify the client connection. It is mapped to a unique pipeline on the Chat Controller server.

user id

unique id to identify the client connection

unique id to identify the client connection

synthesized audio data

The encoding of the audio data

The sample rate in hertz (Hz) of the audio data

The number of channels in the audio data. Only mono is supported

frame size of audio data

unique id to identify the client connection

unique id to identify the client connection

Provides information to the recognizer that specifies how to process the request. The first `SendAudioRequest` message must contain a `streaming_config` message.

The audio data to be recognized. Sequential chunks of audio data are streamed from client.

source id of the audio data

audio buffer creation timestamp in ISO8601 format

Transcript text representing the words that the user spoke.

The non-normalized confidence estimate. A higher number indicates an estimated greater likelihood that the recognized words are correct. This field is set only for a non-streaming result or, of a streaming result where `is_final=true`. This field is not guaranteed to be accurate and users should not rely on it to be always provided.

A list of word-specific information for each recognized word. Only populated if is_final=true

unique id to identify the client connection

Flag to mention whether asr transcripts to be passed to DM-TTS or get only transcripts

The encoding of the audio data sent in the request. All encodings support only 1 channel (mono) audio.

The sample rate in hertz (Hz) of the audio data sent in the `SendAudioRequest` message.

The language of the supplied audio as a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt) language tag. Example: "en-US". Default is en-US.

The number of channels in the input audio data.

Which model to select for the given request.

May contain one or more recognition hypotheses (up to the maximum specified in `max_alternatives`). These alternatives are ordered in terms of accuracy, with the top (first) alternative being the most probable, as ranked by the recognizer.

If `false`, this `StreamingRecognitionResult` represents an interim result that may change. If `true`, this is the final time the speech service will return this particular `StreamingRecognitionResult`, the recognizer will not return any further hypotheses for this portion of the transcript and corresponding audio.

An estimate of the likelihood that the recognizer will not change its guess about this interim result. Values range from 0.0 (completely unstable) to 1.0 (completely stable). This field is only provided for interim results (`is_final=false`). The default of 0.0 is a sentinel value indicating `stability` was not set.

For multi-channel audio, this is the channel number corresponding to the recognized result for the audio from that channel. For audio_channel_count = N, its output values can range from '1' to 'N'.

Length of audio processed so far in seconds

unique id to identify the client connection

uuid to identify concurrent client request

unique id to identify the client connection

message type as defined in `MessageType`

unique id to identify the client connection

transcript text to be synthesized

TTS latency in milliseconds

time in millisecond remained to complete tts audio rendering. This is applicable when tts is set to streaming and realtime from pipeline graph. In non-streaming mode this is expected to be 0.

unique id to identify the client connection

user id

bot name with version like {bot_name}_v{bot_version}, e.g chitchat_bot_v1.

conversation history of user

json formatted data of user context

unique id to identify the client connection

user id

unique id to identify the client connection

used id

bot name with version like {bot_name}_v{bot_version}, e.g chitchat_bot_v1.

Time offset relative to the beginning of the audio in ms and corresponding to the start of the spoken word. This field is only set if `enable_word_time_offsets=true` and only in the top hypothesis.

Time offset relative to the beginning of the audio in ms and corresponding to the end of the spoken word. This field is only set if `enable_word_time_offsets=true` and only in the top hypothesis.

The word corresponding to this set of information.

The non-normalized confidence estimate. A higher number indicates an estimated greater likelihood that the recognized words are correct. This field is not guaranteed to be accurate and users should not rely on it to be always provided. The default of 0.0 is a sentinel value indicating confidence was not set.