This topic describes the Harmony software development kit (SDK) APIs.
API overview
Class/Protocol | API | Description |
ARTCAICallEngineInterface AI call engine interface | Gets the user ID of the current call. | |
Gets the status of whether a call is in progress. | ||
Gets the information of the current agent. | ||
Gets the state of the current agent. | ||
Audio configuration. | ||
Video configuration. | ||
Sets and gets callback events. | ||
Starts a call. | ||
Starts a call using the call configuration. | ||
Hangs up and destroys the current call engine. | ||
Sets the rendering view and configuration for the agent. | ||
Sets the rendering view and configuration for the local camera. | ||
Interrupts the agent's speech. | ||
Sends a text message to the agent. | ||
Sends a custom message to the server. | ||
Updates the system prompt for the large language model (LLM). | ||
Updates the parameters for the Model Studio application center. | ||
Updates the VCR configuration. | ||
Updates the threshold for voice activity detection. | ||
Updates the speech rate for text-to-speech (TTS). | ||
Queries the current state of the agent. | ||
For a Vision agent, starts custom frame capture. | ||
For a Vision agent, stops custom frame capture. | ||
Enables or disables smart interruption. | ||
Switches the voice. | ||
Enables or disables the speaker. | ||
Stops or resumes the playback of the agent's audio stream. | ||
Mutes or unmutes the microphone. | ||
Disables or enables the camera. | ||
Switches between the front and rear cameras. | ||
Enables or disables push-to-talk mode. | ||
Starts speaking. | ||
Speech ended | ||
Cancels the current speech. | ||
Specifies whether to use voiceprint-based denoising for the current speech segment. | ||
Purges the current voiceprint data. | ||
Creates an RTC engine. | ||
Gets the RTC engine. | ||
Releases resources. | ||
Parses shared agent information. | ||
Starts a call with a shared agent. | ||
ARTCAICallEngineListener AI call engine event callback | An error occurred. | |
The connection status changes during a call. | ||
The call starts (joins the session). | ||
The call ends (leaves the session). | ||
The call agent has started. | ||
The RTC engine is created. Call getRTCInstance in this callback to get the RTC engine instance. | ||
Callback for when the agent's message channel is available. Messages can be sent to the agent only after this callback is triggered. | ||
Indicates whether the agent's video stream is available (stream ingest). | ||
Indicates whether the agent's audio stream is available (stream ingest). | ||
The first frame of the agent's digital human is rendered. | ||
The agent's state changes. | ||
The network status changes. | ||
Notification of a volume change. | ||
Notification of the agent's recognition result for a user's question. | ||
Notification of the agent's answer. | ||
Notification of the agent's emotion. | ||
The voice of the current call has changed. | ||
Indicates whether voice interruption is enabled for the current call. | ||
The agent's current speech is interrupted. | ||
Indicates whether push-to-talk mode is enabled for the current call. | ||
Indicates whether voiceprint-based denoising is enabled for the current call. | ||
The voiceprint data for the current call is purged. | ||
The current agent is about to leave (end the current call). | ||
A custom message is received from the current agent. | ||
A VCR result is received from the current agent. | ||
A human is about to take over from the current agent. | ||
A live agent has joined the call. | ||
Audio loopback delay. | ||
Indicates whether custom frame capture mode is enabled for the current Vision call. | ||
The LLM has finished answering in the current call. | ||
If music accompaniment is played through the RTC instance during the current call, this playback callback is triggered. | ||
ARTCAICallEngineFactory Engine factory | Creates a default AICallEngine. |