client.calls

Inbound and outbound calls, their artefacts and live call control. These methods are on client.calls, where client is a MoreVoice client (see the Python SDK). On AsyncMoreVoice the same methods are awaited. Each one returns the response object and raises an exception when the API answers with an error.

add_participant()

Add a participant. Calls handled by a human agent: rings a phone number or an agent into the call (a conference of up to 8), or adds an AI assistant. The participant starts ringing; follow it in GET …/control.

client.calls

def add_participant(self, id: str, body: _m.CallsAddParticipantBody | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallParticipantCreated

client.calls.add_participant() · await async_client.calls.add_participant() · POST /calls/{id}/participants · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
to	object	yes	Who to add: a phone number, an agent or an AI assistant (queues cannot join a call).
announce	string	no	A line said to them when they join.

Returns

CallParticipantCreated:

Field	Type	Description
object	"call_participant"	Always call_participant.
call_id	string	A call ID (prefix call_).
participant_id	string	—
name	string	—
state	"ringing" | "connected"	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_participant_created = client.calls.add_participant("call_7Hk2Lm9Qp", {
    "announce": "Joining you to a customer call",
    "to": {
        "number": "+97235551234",
    },
})
print(call_participant_created)

consult()

Complete, merge, swap or cancel a consultation. After a warm transfer: complete hands the customer to the consulted party, merge makes a three-way call, swap alternates between them, cancel returns to the customer.

client.calls

def consult(self, id: str, action: _m.CallsConsultAction, body: _m.CallsConsultBody | Mapping[str, Any] | Unset = UNSET, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.consult() · await async_client.calls.consult() · POST /calls/{id}/consult/{action} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
action	str	yes	complete: hand the customer over; merge: a three-way call; swap: alternate; cancel: back to the customer.

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.consult("call_7Hk2Lm9Qp", "complete", {})
print(call_control)

create()

Create an outbound call. Places an outbound call with an assistant, or with a published voice flow and its assistant. purpose is required: it drives the outbound compliance checks (do-not-call list for every call; recorded consent and the national registry for marketing). The Idempotency-Key header is required, so a retried request never calls anyone twice. Test-mode keys place simulated calls that never reach a phone network.

client.calls

def create(self, body: _m.CallsCreateBody | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.Call

client.calls.create() · await async_client.calls.create() · POST /calls · API reference

Request body

Field	Type	Required	Description
to	string	yes	The number to call, E.164.
purpose	"service" | "marketing" | "survey"	yes	service, marketing or survey. Marketing calls need recorded consent and pass the national do-not-call registry (§30A).
amd	object	no	Answering-machine detection (default: the assistant’s setting).
assistant_id	string	no	The assistant that handles the call. Pass this or flow_id.
caller_id	string	no	The number to present; it must be allowed on the connection.
connection_id	string	no	The SIP connection to call out on (default: the organisation’s default connection).
flow_id	string	no	A published voice flow; it runs on its linked assistant. Pass this or assistant_id.
from_number_id	string	no	The phone number to call from (pn_…, GET /v1/phone_numbers): it sets the connection and the caller ID. Or pass connection_id and caller_id.
max_duration_s	integer	no	End the call after this many seconds (default: the assistant’s limit).
metadata	MetadataInput	no	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
variables	Record<string, string>	no	Template variables for the prompt, first message and flow ({{customer_name}}); at most 50.

Returns

Call:

Field	Type	Description
id	string	A call ID (prefix call_).
object	"call"	Always call.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
direction	CallDirection | null	—
transport	"sip" | "webrtc"	sip for phone calls, webrtc for browser calls and conference rooms.
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
from	string | null	The calling number (E.164 when it is a phone number) or name.
to	string | null	The called number (E.164 when it is a phone number) or destination.
assistant_id	string | null	asst_… ID.
flow_id	string | null	flow_… ID.
flow_version	integer | null	The flow version that ran (0: an unpublished draft).
queue_id	string | null	q_… ID.
agent_id	string | null	The human agent who handled the call (usr_…).
campaign_id	string | null	cmp_… ID.
contact_id	string | null	ctc_… ID.
connection_id	string | null	conn_… ID.
api_key_id	string | null	The API key that created the call (key_…).
purpose	CallPurpose | null	—
started_at	string	When the call started: dialled out, or rang in.
answered_at	string | null	—
ended_at	string | null	—
duration_ms	integer | null	—
end_reason	string | null	Why the call ended, e.g. customer-ended-call, assistant-ended-call, customer-busy, customer-did-not-answer.
answered_by	CallAnsweredBy | null	—
disposition	string | null	The business outcome recorded on the call (set_outcome tool, or the summary).
has_recording	boolean	—
ai_generated	boolean	Machine-readable marking (EU AI Act Art. 50(2)): true when the call contains speech generated by AI — an AI voice agent, or synthetic-voice IVR prompts. Recordings of such calls carry the same tag in their file metadata.
ai_segments	CallAiSegment[]	When AI-generated speech was played, in ms from the start of the call (empty when ai_generated is false).
summary	CallSummary | null	—
cost	CallCost | null	Set once the call has ended (never null then); null while it is in progress.
variables	Record<string, string>	The variables the call was created with.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
livemode	boolean	true in live mode, false in test mode.
assistant	CallAssistant	The assistant of the call (expand[]=assistant).

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call = client.calls.create({
    "assistant_id": "asst_4Gk2LmN9pQ4rS6tV8wX0yZ",
    "from_number_id": "pn_4Gk2LmN9pQ4rS6tV8wX0yZ",
    "metadata": {
        "crm_contact_id": "0031x00000AbCdE",
    },
    "purpose": "service",
    "to": "+972501234567",
    "variables": {
        "customer_name": "Dana",
    },
})
print(call)

create_stream()

Stream a call’s audio to your WebSocket. Starts a media stream on a live call in fork mode: a one-way copy of the caller (inbound), what the caller hears (outbound), or both, sent to your wss:// URL as Twilio Media Streams messages (connected, start, media, stop). The call goes on exactly as before: if your server is slow, frames older than 2 s are dropped (frames_dropped); if it disconnects, the stream reconnects (3 tries) and otherwise gives up (failed) without touching the call. The upgrade request is signed with your signing secret over its path and query (webhook-id is the stream’s ID). The stream stops when the call ends or with DELETE. At most 2 streams per call.

client.calls

def create_stream(self, id: str, body: _m.MediaStreamCreateParamsInput | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.MediaStream

client.calls.create_stream() · await async_client.calls.create_stream() · POST /calls/{id}/streams · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
url	string	yes	Your WebSocket URL (wss://). The upgrade is signed with your signing secret (webhook-id, webhook-timestamp, webhook-signature over the path and query).
custom_parameters	Record<string, string>	no	Up to 20 string key/values the receiver gets in start.customParameters (as Twilio’s stream parameters).
format	"mulaw_8000" | "l16_16000" | "l16_8000"	no	mulaw_8000 (the default): G.711 μ-law at 8 kHz, Twilio’s; l16_16000 / l16_8000: signed 16-bit PCM, little-endian.
mode	"fork"	no	fork (the default): a one-way copy; your server’s messages are ignored.
tracks	"inbound" | "outbound" | "both"	no	inbound: the caller; outbound: what the caller hears (the assistant, an agent, prompts); both (the default).

Returns

MediaStream:

Field	Type	Description
id	string	The stream’s ID (ms_…): streamSid in the protocol’s messages.
object	"media_stream"	Always media_stream.
livemode	boolean	true in live mode, false in test mode.
call_id	string	The call whose audio is streamed.
mode	"fork"	fork: a one-way copy of the call’s audio; the call goes on as before.
url	string	The receiver’s WebSocket URL.
tracks	"inbound" | "outbound" | "both"	inbound: the caller; outbound: what the caller hears (the assistant, an agent, prompts); both.
format	"mulaw_8000" | "l16_16000" | "l16_8000"	mulaw_8000: G.711 μ-law at 8 kHz (Twilio’s); l16_16000 / l16_8000: signed 16-bit PCM, little-endian.
custom_parameters	Record<string, string>	—
status	"connecting" | "streaming" | "reconnecting" | "stopped" | "failed"	connecting (and reconnecting after the receiver dropped), streaming, stopped, or failed (the receiver could not be reached; the call went on).
frames_sent	integer	20 ms frames sent so far (all tracks).
frames_dropped	integer	Frames dropped because the receiver did not keep up (at most 2 s of audio waits per track).
ended_reason	string | null	Why it ended: call_ended, stopped, receiver_closed or connect_failed; null while it runs.
created	string	An ISO-8601 timestamp in UTC.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

media_stream = client.calls.create_stream("call_7Hk2Lm9Qp", {
    "custom_parameters": {
        "crm_id": "42",
    },
    "format": "mulaw_8000",
    "tracks": "both",
    "url": "wss://media.example.com/streams",
})
print(media_stream)

create_summary()

Summarise a call again. Writes a new AI summary from the transcript and returns it (it replaces the stored one). Send an Idempotency-Key: a retry then returns the same summary instead of writing another.

client.calls

def create_summary(self, id: str, body: _m.CallsCreateSummaryBody | Mapping[str, Any] | Unset = UNSET, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallSummary

client.calls.create_summary() · await async_client.calls.create_summary() · POST /calls/{id}/summary · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

CallSummary:

Field	Type	Description
object	"call_summary"	Always call_summary.
call_id	string	A call ID (prefix call_).
text	string	A two-to-four sentence summary, in the language of the call.
intent	string	What the caller wanted.
outcome	string	How the call ended, or what was agreed.
sentiment	"positive" | "neutral" | "negative" | "mixed"	—
key_points	string[]	—
action_items	string[]	Follow-ups for your business; empty when there are none.
generated_at	string | null	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_summary = client.calls.create_summary("call_7Hk2Lm9Qp", {})
print(call_summary)

delete()

Delete a call. Deletes an ended call with its transcript, events and recording. A call in progress answers 409 call_in_progress: hang it up first.

client.calls

def delete(self, id: str, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallDeleted

client.calls.delete() · await async_client.calls.delete() · DELETE /calls/{id} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

CallDeleted:

Field	Type	Description
id	string	A call ID (prefix call_).
object	"call"	Always call.
deleted	true	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_deleted = client.calls.delete("call_7Hk2Lm9Qp")
print(call_deleted)

delete_stream()

Stop a media stream. Stops the stream: what is queued is flushed, your server gets stop, and the socket closes. The call is not affected. Streams also stop on their own when the call ends.

client.calls

def delete_stream(self, id: str, stream_id: str, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.MediaStream

client.calls.delete_stream() · await async_client.calls.delete_stream() · DELETE /calls/{id}/streams/{stream_id} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
stream_id	str	yes	The stream’s ID (ms_…).

Returns

MediaStream:

Field	Type	Description
id	string	The stream’s ID (ms_…): streamSid in the protocol’s messages.
object	"media_stream"	Always media_stream.
livemode	boolean	true in live mode, false in test mode.
call_id	string	The call whose audio is streamed.
mode	"fork"	fork: a one-way copy of the call’s audio; the call goes on as before.
url	string	The receiver’s WebSocket URL.
tracks	"inbound" | "outbound" | "both"	inbound: the caller; outbound: what the caller hears (the assistant, an agent, prompts); both.
format	"mulaw_8000" | "l16_16000" | "l16_8000"	mulaw_8000: G.711 μ-law at 8 kHz (Twilio’s); l16_16000 / l16_8000: signed 16-bit PCM, little-endian.
custom_parameters	Record<string, string>	—
status	"connecting" | "streaming" | "reconnecting" | "stopped" | "failed"	connecting (and reconnecting after the receiver dropped), streaming, stopped, or failed (the receiver could not be reached; the call went on).
frames_sent	integer	20 ms frames sent so far (all tracks).
frames_dropped	integer	Frames dropped because the receiver did not keep up (at most 2 s of audio waits per track).
ended_reason	string | null	Why it ended: call_ended, stopped, receiver_closed or connect_failed; null while it runs.
created	string	An ISO-8601 timestamp in UTC.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

media_stream = client.calls.delete_stream("call_7Hk2Lm9Qp", "ms_4fG7hJ2kL9mN3pQ6rS8tUv")
print(media_stream)

download_recording()

Download a recording (signed link). The URL a recording link points at. No API key: the token is the credential, valid for one call for 15 minutes. Supports Range requests.

client.calls

def download_recording(self, token: str) -> bytes

client.calls.download_recording() · await async_client.calls.download_recording() · GET /recordings/{token} · API reference

Parameters

Name	Type	Required	Description
token	str	yes	The signed token of a recording link.

Returns

Nothing, on success.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

data = client.calls.download_recording("token_123")
print(len(data), "bytes")

hangup()

Hang up a call. Ends a live call for everyone on it (while it rings, cancels it) and returns the call. An ended call answers 409 call_not_active.

client.calls

def hangup(self, id: str, body: _m.CallsHangupBody | Mapping[str, Any] | Unset = UNSET, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.Call

client.calls.hangup() · await async_client.calls.hangup() · POST /calls/{id}/hangup · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

Call:

Field	Type	Description
id	string	A call ID (prefix call_).
object	"call"	Always call.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
direction	CallDirection | null	—
transport	"sip" | "webrtc"	sip for phone calls, webrtc for browser calls and conference rooms.
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
from	string | null	The calling number (E.164 when it is a phone number) or name.
to	string | null	The called number (E.164 when it is a phone number) or destination.
assistant_id	string | null	asst_… ID.
flow_id	string | null	flow_… ID.
flow_version	integer | null	The flow version that ran (0: an unpublished draft).
queue_id	string | null	q_… ID.
agent_id	string | null	The human agent who handled the call (usr_…).
campaign_id	string | null	cmp_… ID.
contact_id	string | null	ctc_… ID.
connection_id	string | null	conn_… ID.
api_key_id	string | null	The API key that created the call (key_…).
purpose	CallPurpose | null	—
started_at	string	When the call started: dialled out, or rang in.
answered_at	string | null	—
ended_at	string | null	—
duration_ms	integer | null	—
end_reason	string | null	Why the call ended, e.g. customer-ended-call, assistant-ended-call, customer-busy, customer-did-not-answer.
answered_by	CallAnsweredBy | null	—
disposition	string | null	The business outcome recorded on the call (set_outcome tool, or the summary).
has_recording	boolean	—
ai_generated	boolean	Machine-readable marking (EU AI Act Art. 50(2)): true when the call contains speech generated by AI — an AI voice agent, or synthetic-voice IVR prompts. Recordings of such calls carry the same tag in their file metadata.
ai_segments	CallAiSegment[]	When AI-generated speech was played, in ms from the start of the call (empty when ai_generated is false).
summary	CallSummary | null	—
cost	CallCost | null	Set once the call has ended (never null then); null while it is in progress.
variables	Record<string, string>	The variables the call was created with.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
livemode	boolean	true in live mode, false in test mode.
assistant	CallAssistant	The assistant of the call (expand[]=assistant).

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call = client.calls.hangup("call_7Hk2Lm9Qp", {})
print(call)

hold()

Put a call on hold or take it off hold. Calls handled by a human agent: the customer hears the hold music while on is true.

client.calls

def hold(self, id: str, body: _m.CallsHoldBody | Mapping[str, Any] | Unset = UNSET, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.hold() · await async_client.calls.hold() · POST /calls/{id}/hold · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
on	boolean	no	true: put the customer on hold; false: take them off hold.

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.hold("call_7Hk2Lm9Qp", {
    "on": True,
})
print(call_control)

list()

List calls. The organisation’s calls, newest first. Test-mode keys see only test calls. Filter by status, direction, type, assistant, campaign, agent, number or start time (created[gte], created[lt]…).

client.calls

def list(self, *, limit: int | Unset = 20, starting_after: str | Unset = UNSET, ending_before: str | Unset = UNSET, status: _m.CallsListStatus | Unset = UNSET, direction: _m.CallsListDirection | Unset = UNSET, type_: _m.CallsListType | Unset = UNSET, assistant_id: str | Unset = UNSET, campaign_id: str | Unset = UNSET, agent_id: str | Unset = UNSET, from_: str | Unset = UNSET, to: str | Unset = UNSET, created: _m.CallsListCreated | Unset = UNSET, expand: _b.list[str] | Unset = UNSET, more_voice_version: str | Unset = UNSET) -> AsyncPage[_m.Call]

client.calls.list() · await async_client.calls.list() · GET /calls · API reference

Parameters

Name	Type	Required	Description
limit	integer	no	How many objects to return, 1–100 (default 20).
starting_after	string	no	A cursor (next_cursor) or object ID: return the objects after it (older).
ending_before	string	no	A cursor or object ID: return the objects before it (newer).
status	"in_progress" | "ended"	no	in_progress: calls that have not ended (queued, ringing or answered). ended: finished calls.
direction	"inbound" | "outbound" | "browser"	no
type	"ai" | "human" | "ivr" | "conference" | "voicemail"	no
assistant_id	string	no	A assistant ID (asst_…).
campaign_id	string	no	A campaign ID (cmp_…).
agent_id	string	no	A user ID (usr_…).
from	string	no	Calls from this number, exactly as recorded on the call.
to	string	no	Calls to this number, exactly as recorded on the call.
created	object	no	Filter on started_at: created[gte], created[gt], created[lte], created[lt] (ISO-8601).
expand	string[]	no	Fields to expand into objects, e.g. expand[]=assistant.

Returns

A page of results (CallList): data, has_more and next_cursor.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

for call in client.calls.list():
    print(call)

list_copilot_events()

List a call’s copilot events. What the real-time agent copilot showed on the call and what the agent did with it (cards used or dismissed, searches), newest first.

client.calls

def list_copilot_events(self, id: str, *, limit: int | Unset = 20, starting_after: str | Unset = UNSET, ending_before: str | Unset = UNSET, more_voice_version: str | Unset = UNSET) -> AsyncPage[_m.CopilotEvent]

client.calls.list_copilot_events() · await async_client.calls.list_copilot_events() · GET /calls/{id}/copilot_events · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
limit	integer	no	How many objects to return, 1–100 (default 20).
starting_after	string	no	A cursor (next_cursor) or object ID: return the objects after it (older).
ending_before	string	no	A cursor or object ID: return the objects before it (newer).

Returns

A page of results (CopilotEventList): data, has_more and next_cursor.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

for call in client.calls.list_copilot_events("call_7Hk2Lm9Qp"):
    print(call)

pci()

Pause or resume recording for a card payment. PCI DSS scope reduction: while paused the call is neither recorded nor transcribed, live listeners hear silence and keypad digits are logged masked; with DTMF masking on, keypad tones are removed for everyone except the payment line. Pausing a paused call (or resuming a running one) is a 409. Needs the organisation’s PCI mode.

client.calls

def pci(self, id: str, body: _m.CallsPciBody | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallPci

client.calls.pci() · await async_client.calls.pci() · POST /calls/{id}/pci · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
action	"pause" | "resume"	yes	pause: stop recording and transcription for a card payment; resume: start them again.
destination_participant_id	string	no	Human-handled calls: the participant (a payment IVR or keypad-capture line you added) that receives the caller’s keys.
max_seconds	integer	no	Resume by itself after this long (capped by the organisation’s maximum, which is also the default).
reason	string	no	Shown in the call log (default payment).

Returns

CallPci:

Field	Type	Description
object	"call_pci"	Always call_pci.
call_id	string	A call ID (prefix call_).
paused	boolean	Recording and transcription are paused for a payment.
reason	string | null	Why (e.g. payment), while paused.
paused_at	string | null	—
resumes_at	string | null	When the pause ends by itself unless resumed earlier.
destination_participant_id	string | null	The participant the pause named as the payment line, if any (otherwise a participant on one of the organisation’s payment numbers).

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_pci = client.calls.pci("call_7Hk2Lm9Qp", {
    "action": "pause",
    "max_seconds": 180,
    "reason": "payment",
})
print(call_pci)

remove_participant()

Remove a participant. Drops a participant from the call. The customer (client) cannot be removed: hang up instead.

client.calls

def remove_participant(self, id: str, participant_id: str, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.remove_participant() · await async_client.calls.remove_participant() · DELETE /calls/{id}/participants/{participant_id} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
participant_id	str	yes	A participant_id from the call’s control state.

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.remove_participant("call_7Hk2Lm9Qp", "participant_id_123")
print(call_control)

retrieve()

Retrieve a call. A call, live or ended. While it runs, status and answered_at come from the live call.

client.calls

def retrieve(self, id: str, *, expand: _b.list[str] | Unset = UNSET, more_voice_version: str | Unset = UNSET) -> _m.Call

client.calls.retrieve() · await async_client.calls.retrieve() · GET /calls/{id} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
expand	string[]	no	Fields to expand into objects, e.g. expand[]=assistant.

Returns

Call:

Field	Type	Description
id	string	A call ID (prefix call_).
object	"call"	Always call.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
direction	CallDirection | null	—
transport	"sip" | "webrtc"	sip for phone calls, webrtc for browser calls and conference rooms.
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
from	string | null	The calling number (E.164 when it is a phone number) or name.
to	string | null	The called number (E.164 when it is a phone number) or destination.
assistant_id	string | null	asst_… ID.
flow_id	string | null	flow_… ID.
flow_version	integer | null	The flow version that ran (0: an unpublished draft).
queue_id	string | null	q_… ID.
agent_id	string | null	The human agent who handled the call (usr_…).
campaign_id	string | null	cmp_… ID.
contact_id	string | null	ctc_… ID.
connection_id	string | null	conn_… ID.
api_key_id	string | null	The API key that created the call (key_…).
purpose	CallPurpose | null	—
started_at	string	When the call started: dialled out, or rang in.
answered_at	string | null	—
ended_at	string | null	—
duration_ms	integer | null	—
end_reason	string | null	Why the call ended, e.g. customer-ended-call, assistant-ended-call, customer-busy, customer-did-not-answer.
answered_by	CallAnsweredBy | null	—
disposition	string | null	The business outcome recorded on the call (set_outcome tool, or the summary).
has_recording	boolean	—
ai_generated	boolean	Machine-readable marking (EU AI Act Art. 50(2)): true when the call contains speech generated by AI — an AI voice agent, or synthetic-voice IVR prompts. Recordings of such calls carry the same tag in their file metadata.
ai_segments	CallAiSegment[]	When AI-generated speech was played, in ms from the start of the call (empty when ai_generated is false).
summary	CallSummary | null	—
cost	CallCost | null	Set once the call has ended (never null then); null while it is in progress.
variables	Record<string, string>	The variables the call was created with.
metadata	Metadata	Up to 50 key/value pairs (keys up to 40 characters, values up to 500) you attach to an object. Returned as sent.
livemode	boolean	true in live mode, false in test mode.
assistant	CallAssistant	The assistant of the call (expand[]=assistant).

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call = client.calls.retrieve("call_7Hk2Lm9Qp")
print(call)

retrieve_control()

Retrieve a call’s control state. Who is on the call, hold and consultation state. Ended calls answer status ended with no one connected.

client.calls

def retrieve_control(self, id: str, *, more_voice_version: str | Unset = UNSET) -> _m.CallControl

client.calls.retrieve_control() · await async_client.calls.retrieve_control() · GET /calls/{id}/control · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.retrieve_control("call_7Hk2Lm9Qp")
print(call_control)

retrieve_flow_path()

Retrieve a call’s flow path. The flow nodes the call went through, in order, with the edge taken into each. Calls without a flow answer flow_id: null and no steps.

client.calls

def retrieve_flow_path(self, id: str, *, more_voice_version: str | Unset = UNSET) -> _m.FlowPath

client.calls.retrieve_flow_path() · await async_client.calls.retrieve_flow_path() · GET /calls/{id}/flow_path · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

FlowPath:

Field	Type	Description
object	"flow_path"	Always flow_path.
call_id	string	A call ID (prefix call_).
flow_id	string | null	null when no flow ran on the call.
flow_version	integer | null	—
steps	FlowPathStep[]	—
end	string | null	How the flow ended (an end node, a transfer, a hang-up).
transferred	object | null	—
live	boolean	The call is still running: more steps may follow.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

flow_path = client.calls.retrieve_flow_path("call_7Hk2Lm9Qp")
print(flow_path)

retrieve_pci()

Retrieve a call’s payment pause. Returns the CallPci object. Answers 404 with the code resource_missing when nothing has this ID in this organisation and mode.

client.calls

def retrieve_pci(self, id: str, *, more_voice_version: str | Unset = UNSET) -> _m.CallPci

client.calls.retrieve_pci() · await async_client.calls.retrieve_pci() · GET /calls/{id}/pci · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

CallPci:

Field	Type	Description
object	"call_pci"	Always call_pci.
call_id	string	A call ID (prefix call_).
paused	boolean	Recording and transcription are paused for a payment.
reason	string | null	Why (e.g. payment), while paused.
paused_at	string | null	—
resumes_at	string | null	When the pause ends by itself unless resumed earlier.
destination_participant_id	string | null	The participant the pause named as the payment line, if any (otherwise a participant on one of the organisation’s payment numbers).

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_pci = client.calls.retrieve_pci("call_7Hk2Lm9Qp")
print(call_pci)

retrieve_recording()

Retrieve a call’s recording. Redirects (302) to a signed URL that streams the recording for 15 minutes, with no API key. With format=json, answers that URL as a recording_link object instead. The recording is stereo: the customer on the left channel, the assistant or agent on the right.

client.calls

def retrieve_recording(self, id: str, *, format_: _m.CallsRetrieveRecordingFormat | Unset = "redirect", more_voice_version: str | Unset = UNSET) -> _m.RecordingLink

client.calls.retrieve_recording() · await async_client.calls.retrieve_recording() · GET /calls/{id}/recording · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
format	"redirect" | "json"	no	redirect (default): 302 to the signed URL. json: the signed URL as a recording_link object.

Returns

RecordingLink:

Field	Type	Description
object	"recording_link"	Always recording_link.
call_id	string	A call ID (prefix call_).
url	string	A signed URL that streams the audio without an API key. It expires (expires_at).
expires_at	string	An ISO-8601 timestamp in UTC.
content_type	string	audio/wav, audio/ogg, audio/webm or audio/mpeg.
duration_ms	integer | null	—
channels	integer	2: the customer on the left channel, the assistant or agent on the right.

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

recording_link = client.calls.retrieve_recording("call_7Hk2Lm9Qp")
print(recording_link)

retrieve_summary()

Retrieve a call’s summary. The AI summary written when the call ended (when the assistant’s summary artefact is on), or the last one created with POST.

client.calls

def retrieve_summary(self, id: str, *, more_voice_version: str | Unset = UNSET) -> _m.CallSummary

client.calls.retrieve_summary() · await async_client.calls.retrieve_summary() · GET /calls/{id}/summary · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

CallSummary:

Field	Type	Description
object	"call_summary"	Always call_summary.
call_id	string	A call ID (prefix call_).
text	string	A two-to-four sentence summary, in the language of the call.
intent	string	What the caller wanted.
outcome	string	How the call ended, or what was agreed.
sentiment	"positive" | "neutral" | "negative" | "mixed"	—
key_points	string[]	—
action_items	string[]	Follow-ups for your business; empty when there are none.
generated_at	string | null	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_summary = client.calls.retrieve_summary("call_7Hk2Lm9Qp")
print(call_summary)

retrieve_transcript()

Retrieve a call’s transcript. Everything said on the call, in order, with the speaker of each segment. While the call runs, the transcript so far.

client.calls

def retrieve_transcript(self, id: str, *, more_voice_version: str | Unset = UNSET) -> _m.Transcript

client.calls.retrieve_transcript() · await async_client.calls.retrieve_transcript() · GET /calls/{id}/transcript · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Returns

Transcript:

Field	Type	Description
object	"transcript"	Always transcript.
call_id	string	A call ID (prefix call_).
language	string | null	The call’s main language (BCP 47), when known.
segments	TranscriptSegment[]	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

transcript = client.calls.retrieve_transcript("call_7Hk2Lm9Qp")
print(transcript)

send_dtmf()

Send DTMF tones. Calls handled by a human agent: plays the keys to the far end (for example to navigate another company’s IVR).

client.calls

def send_dtmf(self, id: str, body: _m.CallsSendDtmfBody | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.send_dtmf() · await async_client.calls.send_dtmf() · POST /calls/{id}/dtmf · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
digits	string	yes	The keys to send to the far end; , pauses.

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.send_dtmf("call_7Hk2Lm9Qp", {
    "digits": "1234#",
})
print(call_control)

stream()

Stream a call’s live events. Server-Sent Events for one call: call.state_changed, transcript.partial, transcript.final, call.tool_called, and call.ended, after which the stream closes. The stream starts with what already happened on the call. Reconnect with Last-Event-ID (EventSource does it for you) to get only what you missed. An ended call’s stream sends call.ended at once. A call that has not started yet (a web call before the browser connects) is a 404: subscribe once it starts, or follow it on GET /v1/events/stream. Counts against the key’s stream limit (concurrency_limit, limit_type: api_streams).

client.calls.stream() · await async_client.calls.stream() · GET /calls/{id}/events · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
types	string | string[]	no	Only these event types (comma-separated or repeated): call.state_changed, transcript.partial, transcript.final, call.tool_called, call.ended, transcript.*, call.* or *. Default: all of them.
last_event_id	string	no	Resume after this event, for clients that cannot send the Last-Event-ID header (the header wins when both are sent).

Returns

A stream of events (Server-Sent Events): iterate it, or async for it on AsyncMoreVoice.

Example

import os

import httpx

url = "https://api.morevoice.ai/v1/calls/call_7Hk2Lm9Qp/events"
headers = {"Authorization": f"Bearer {os.environ['MOREVOICE_API_KEY']}", "Accept": "text/event-stream"}
with httpx.stream("GET", url, headers=headers, timeout=None) as response:
    response.raise_for_status()
    for line in response.iter_lines():
        if line.startswith("data:"):
            print(line[5:].strip())

transcript_stream()

Stream a call’s live transcript. Server-Sent Events with what is said on a call as it is said: transcript.partial while a line is spoken (its text so far) and transcript.final when it is complete. A final supersedes the partials with the same utterance_id. speaker is customer (the caller), assistant (the AI), agent (a human agent), supervisor or external (a transfer party); AI and human calls alike. start_ms and end_ms are when the line began and became final, in milliseconds from the call’s start, as observed live (null for lines spoken before the stream connected; the call’s transcript has the exact times after the call). The stream starts with the conversation so far and ends with call.ended. Reconnect with Last-Event-ID to get only what you missed. Send partials=false for complete lines only. Counts against the key’s stream limit (concurrency_limit, limit_type: api_streams).

client.calls.transcript_stream() · await async_client.calls.transcript_stream() · GET /calls/{id}/transcript/stream · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
partials	"true" | "false"	no	false: only complete lines (transcript.final), no partials. Default true.
last_event_id	string	no	Resume after this event, for clients that cannot send the Last-Event-ID header (the header wins when both are sent).

Returns

A stream of events (Server-Sent Events): iterate it, or async for it on AsyncMoreVoice.

Example

import os

import httpx

url = "https://api.morevoice.ai/v1/calls/call_7Hk2Lm9Qp/transcript/stream"
headers = {"Authorization": f"Bearer {os.environ['MOREVOICE_API_KEY']}", "Accept": "text/event-stream"}
with httpx.stream("GET", url, headers=headers, timeout=None) as response:
    response.raise_for_status()
    for line in response.iter_lines():
        if line.startswith("data:"):
            print(line[5:].strip())

transfer()

Transfer a call. AI and IVR calls: to a queue or a phone number (warm: note is whispered to the person who answers). Calls handled by a human agent: to a queue, a number, an AI assistant or another agent, cold or warm (warm: the agent consults the target first; finish with POST …/consult/complete). Returns the call’s control state.

client.calls

def transfer(self, id: str, body: _m.CallsTransferBody | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.transfer() · await async_client.calls.transfer() · POST /calls/{id}/transfer · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).

Request body

Field	Type	Required	Description
to	object	yes	Where to: a queue, a phone number, an AI assistant or a human agent.
mode	"cold" | "warm"	no	cold: hand the call over. warm: talk to the target first (human calls), or whisper note to them (AI calls).
note	string	no	Context for the target: shown with the queued call, whispered on a warm transfer.

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.transfer("call_7Hk2Lm9Qp", {
    "mode": "warm",
    "note": "VIP customer, order A-1042",
    "to": {
        "number": "+97235551234",
    },
})
print(call_control)

update_participant()

Mute, hold or change the role of a participant. Changes one participant of a call handled by a human agent: mute or unmute them (muted), put them on hold or take them off (held), or change their role. Send only what changes. The answer is the call’s control state after the change. AI and IVR calls answer 400 with the code unsupported_for_call_mode.

client.calls

def update_participant(self, id: str, participant_id: str, body: _m.CallsUpdateParticipantBody | Mapping[str, Any] | Unset = UNSET, *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.CallControl

client.calls.update_participant() · await async_client.calls.update_participant() · PATCH /calls/{id}/participants/{participant_id} · API reference

Parameters

Name	Type	Required	Description
id	str	yes	A call ID (call_…).
participant_id	str	yes	A participant_id from the call’s control state.

Request body

Field	Type	Required	Description
held	boolean	no	—
muted	boolean	no	—
role	"moderator" | "member"	no	—

Returns

CallControl:

Field	Type	Description
object	"call_control"	Always call_control.
call_id	string	A call ID (prefix call_).
status	CallStatus	queued (placed, not ringing yet), ringing, in_progress (answered, or an inbound call being handled) or ended.
type	CallType	Who handles the call: an AI assistant, a human agent, an IVR flow, a conference room, or a voicemail box.
phase	"active" | "held" | "consulting" | "conference"	—
held	boolean	The customer is on hold.
consult	object | null	A running consultation (warm transfer).
participants	CallParticipant[]	—
recording	boolean	—

Example

from morevoice import MoreVoice

client = MoreVoice()  # MOREVOICE_API_KEY from the environment

call_control = client.calls.update_participant("call_7Hk2Lm9Qp", "participant_id_123", {
    "muted": True,
})
print(call_control)