Skip to main content
The action defines what the voice agent will do during the call with your systems. This includes the tasks to be accomplished in response to user requests.
An agent is an intelligent system capable of handling incoming and outgoing calls, using advanced language models to interact smoothly with users.
The API (Application Programming Interface) allows developers to integrate and communicate with the Volubile platform, facilitating access to the voice agent’s features.
A campaign is an organized set of outgoing calls, often targeted towards a specific group of users, aiming to achieve marketing or communication objectives.
The classifier groups the categories in which the completed call will be classified, allowing for informed decisions about the further treatment of the interaction.
Context encompasses the information provided during the call, such as the caller’s first and last name, as well as interaction history, enriching the conversation.
DTMF (Dual-Tone Multi-Frequency) is a telephone signaling system that uses tones to transmit digits and commands during a call. This technology allows users to interact with the voice agent by pressing the keys on their phone.
The extractor refers to the elements we wish to automatically extract from the call, such as the first name, last name, order number, etc., for further analysis.
A homophone is a word that sounds the same as another word but has a different meaning, which can pose challenges in voice recognition.
An incoming call refers to a telephone call received by the system, which will be managed by the voice agent to provide assistance or information to the callers.
IVR (Interactive Voice Response) is an automated telephone system that allows callers to navigate a voice menu and access specific information or services. Volubile agents can integrate and enhance the functionalities of traditional IVR systems.
LLM (Large Language Model) refers to large-scale language models that form the technological core of Volubile voice agents. These artificial intelligence models enable understanding, processing, and generating natural responses during phone interactions.
An outgoing call refers to a telephone call initiated by the system to contact a user or a client, often used in marketing or follow-up campaigns.
A phoneme is the smallest unit of sound in a language that can change the meaning of a word. Voice recognition and synthesis rely on phoneme analysis for effective communication.
The prompt is a behavioral element of the AI agent that guides the conversation. It determines how the agent interacts with the user, influencing the flow of the exchange and the way responses are formulated.
RAG (Retrieval-Augmented Generation) is a method that combines information retrieval and text generation to improve the accuracy and relevance of the responses provided by the agent.
SIP (Session Initiation Protocol) is a communication protocol used to establish, manage, and terminate call sessions over the Internet, essential for the operation of voice agents.
Speech recognition is the technology that converts spoken words into text, enabling the agent to understand and interpret users’ voice commands.
A tag is a label assigned to calls or interactions, which facilitates tracking, organizing, and analyzing data within the system.
Text-to-speech is the technology that converts text into spoken words, allowing the agent to communicate effectively with users in a natural and fluent manner.
TruePath is a proprietary technology from Volubile that enables conversational management without hallucinations, ensuring precise and relevant interactions.
A webhook is an end-of-call notification sent to your systems, containing key elements for further processing or data updates.
A workflow is a process that allows, from the call, to take actions on your systems based on the classifiers and extractors, ensuring effective integration.