Make and control calls

Place outbound calls to your agent, transfer a live call to a human, play or record, and start or stop a stream on a live call.

Place an outbound call, inspect it, or control its audio while it is active. Log in, use a caller number on your account, and choose a test recipient:

plivo voice calls make --from [phone] --to [phone] --dry-run

This previews a call that plays Plivo’s hello demo. Replace --dry-run with --yes to place the billed call. To call your voice agent instead, supply its public answer URL, which returns <Stream> XML:

plivo voice calls make --from [phone] --to [phone] \
  --answer-url https://agent.example.com/answer --answer-method POST --dry-run

For human handoff, run plivo voice calls transfer "<call_uuid>" --legs aleg --aleg-url "<url>", where the new URL returns <Dial> XML.

• Find a call UUID with plivo voice calls list after the call ends, or plivo voice calls list --live while it is active. A call leg is one side of a connected call (aleg is the caller, bleg the callee). DTMF means keypad tones, and TTS means text-to-speech.
• Replace example numbers, URLs and <values> with yours. Omit [flags] unless adding options.
• Global flags: see Configuration.

​

Calls

voice calls make spends money and requires --yes. Live-call controls (play, speak, record, dtmf, transfer, stop-*) act on an established call and don’t need it. hangup needs --yes, and previews with --dry-run alone. With -o json, the live-call controls and hangup print a result on stdout, such as {"data": {"call_uuid": "<call_uuid>", "hung_up": true}}.

​

Read a call in progress

voice calls get reads the call record, which Plivo writes when the call ends. Add --live to read a call still in progress. If the call has already ended, --live returns not found.

voice calls list --live lists the calls in progress. Plivo returns only their UUIDs, with no filter or paging, so --live takes no other list flag. With -o json the UUIDs are in data.calls.

plivo voice calls get "<call_uuid>" --live
plivo voice calls list --live

​

Filter the call list by date

plivo voice calls list shows the newest calls first. Without a date, Plivo returns the last 7 days.

plivo voice calls list --since 2026-10-01 --until 2026-10-07
plivo voice calls list --since "2026-10-01 09:00" --until 2026-10-07 -o json
plivo voice calls list --since 2026-10-01 --until 2026-10-07 --all          # every page, not just the first 20 rows

• --since and --until filter on when a call ended, in UTC, as YYYY-MM-DD or YYYY-MM-DD HH:MM[:SS]. --from-date and --to-date are the same flags.
• A bare --until date includes that whole day. --until needs --since.
• Plivo searches only the last 90 days, at most 31 days per request. A range outside either limit is refused before any request, and the error says what to run instead, such as the ranges to split it into.

Usage	What it does
plivo voice calls make [flags]	Make an outbound call (requires —yes; spends money: use —dry-run to preview)
plivo voice calls list [flags]	List calls (paged with --limit, --offset, --all)
plivo voice calls get <call_uuid> [flags]	Get a call by UUID. --live gets a call in progress, not its record.
plivo voice calls hangup <call_uuid>	Hang up a live call (requires —yes)
plivo voice calls transfer <call_uuid> [flags]	Transfer one or both legs of a live call to new PlivoXML URLs
plivo voice calls play <call_uuid> [flags]	Play audio file(s) into a live call
plivo voice calls stop-play <call_uuid>	Stop any audio playing in a live call
plivo voice calls speak <call_uuid> [flags]	Speak text into a live call via TTS
plivo voice calls stop-speak <call_uuid>	Stop any TTS playing in a live call
plivo voice calls record <call_uuid> [flags]	Start recording a live call
plivo voice calls stop-record <call_uuid>	Stop the active recording on a live call
plivo voice calls dtmf <call_uuid> [flags]	Send DTMF tones into a live call

plivo voice calls make

Flag	Description
--answer-method <value>	GET|POST (default: GET for the hello demo, else Plivo’s POST)
--answer-url <value>	URL returning PlivoXML to play on answer (default: Plivo’s hello demo) (default “https://s3.amazonaws.com/static.plivo.com/answer.xml”)
--explain	narrate in plain English before executing
--from <value>	source number in international format, such as [phone]; must be on your account (required)
--hangup-url <value>	URL hit when call ends
--machine-detection <value>	answering-machine detection: true|hangup (off when omitted)
--ring-url <value>	URL hit when call starts ringing
--to <value>	destination: an international number with country code or a sip: URI; quote multiple destinations joined by < (required)

A --from that isn’t an E.164 number, or a --to part that is neither an E.164 number nor a sip: URI, is refused before any request.

plivo voice calls list

Flag	Description
--direction <value>	inbound|outbound
--from <value>	filter by from_number
--to <value>	filter by to_number
--live	list the UUIDs of the calls in progress, not call records (takes no other list flag)
--since <value>	calls ending at or after this UTC time (YYYY-MM-DD[ HH:MM[:SS]]), within the last 90 days, at most 31 days per request
--until <value>	calls ending at or before this UTC time (YYYY-MM-DD[ HH:MM[:SS]])
--from-date <value>	same as —since
--to-date <value>	same as —until

plivo voice calls transfer

Flag	Description
--aleg-method <value>	GET|POST (default “POST”)
--aleg-url <value>	new URL for A-leg (caller side)
--bleg-method <value>	GET|POST (default “POST”)
--bleg-url <value>	new URL for B-leg (callee side)
--legs <value>	which leg(s) to act on: aleg|bleg|both (default “aleg”)

plivo voice calls play

Flag	Description
--legs <value>	aleg|bleg|both (default “aleg”)
--length <integer>	stop after N seconds (0 = full file)
--loop	loop the playback
--mix	mix with the call audio (else replace) (default true)
--urls <value>	comma-separated list of audio file URLs to play (required)

plivo voice calls speak

Flag	Description
--language <value>	language code, e.g. en-US, en-GB, hi-IN (default “en-US”)
--legs <value>	aleg|bleg|both (default “aleg”)
--mix	mix with the call audio (else replace) (default true)
--text <value>	text to speak via TTS (required)
--voice <value>	voice: MAN|WOMAN (default “WOMAN”)

plivo voice calls record

Flag	Description
--callback-method <value>	GET|POST (default “POST”)
--callback-url <value>	URL hit when recording finishes
--file-format <value>	mp3|wav (default “mp3”)
--record-channel-type <value>	mono|stereo (stereo puts each leg on its own channel)
--time-limit <integer>	max recording length in seconds (default 60)
--transcription-type <value>	transcription type: auto|hybrid|manual (results go to —transcription-url)
--transcription-url <value>	URL the transcription is sent to

plivo voice calls dtmf

Flag	Description
--digits <value>	DTMF digits to send, e.g. 1234# (required)
--leg <value>	aleg|bleg (default “aleg”)

​

Streams on a live call (REST)

Start a WebSocket audio stream on an already-answered call (instead of via <Stream> XML), list the streams on a call, or stop them.

Usage	What it does
plivo voice calls streams start <call_uuid> [flags]	Start a new audio stream on a call
plivo voice calls streams list <call_uuid>	List active streams on a call
plivo voice calls streams get <call_uuid> <stream_id>	Get details for one stream on a call
plivo voice calls streams stop <call_uuid> [stream_id]	Stop one stream (with id) or all streams (without id) on a call

plivo voice calls streams start

Flag	Description
--audio-track <value>	inbound|outbound|both (default “inbound”)
--bidirectional	let the WebSocket send audio back into the call (needs —audio-track inbound)
--callback-url <value>	alias for —stream-status-callback
--content-type <value>	audio codec content-type (default “audio/x-l16;rate=16000”)
--extra-headers <value>	comma-separated extra WebSocket headers (k1=v1,k2=v2)
--stream-status-callback <value>	URL hit when stream starts/ends
--url <value>	WebSocket URL to receive audio (wss://…) (required)