AI agents for Cartesia

List voice agents

Lists the organization's Line voice agents in summaries: each with id, name, description, voice, language, phone_numbers (id and number) and dates.

List calls

Lists one agent's calls (or one outbound batch's calls), newest first.

Get a call

Reads one call: status, start and end times, end_reason, the Cartesia-written summary, the caller and callee in telephony_params, the transcript turn by turn.

List phone numbers

Lists the organization's phone numbers: each with id (pn_…, the from_number_id for outbound calls), number, label and the agent it is assigned to.

Get an outbound batch

Reads an outbound call batch: status and how many calls were scheduled, dispatched and finished, with each recipient's state.

List knowledge base documents

Lists the knowledge base documents voice agents answer from: each with id, name, content, metadata and created_at.

Transcribe an audio file

Transcribes an audio file (a person's attachment or another tool's answer, passed as its file reference fil_…; flac, m4a, mp3, mp4, ogg, wav or webm, at most 25 MB here) with Cartesia's Ink Whisper…

List voices

Searches Cartesia's voices and the organization's own: each with id, name, tagline, description, gender, accents and whether you own it.

Start outbound calls

Queues real phone calls from a voice agent to one or more numbers (E.164, like +14155559876), each with its own dynamic variables (like the person's name or order number), now or at scheduled_at.

Add a knowledge base document

Adds a text document (up to 1 MB, like a return policy, price list or FAQ) to a knowledge base folder, so the agents with that folder can answer from it.

Update a knowledge base document

Replaces a knowledge base document's text, name or metadata, or moves it to another folder.

Turn text into speech

Speaks a text with one of Cartesia's voices (find one with list_voices) and returns the audio as a file reference in file: MP3 by default, or WAV.