Automated voice

Text-to-speech voice routes for alerts and mass notification

Turn a line of text into thousands of simultaneous calls, with natural-sounding output, concurrency sized to your window and per-call disposition you can act on.

One text payload, thousands of concurrent calls
Natural-sounding voices, or bring your own recording
Concurrency provisioned for your campaign window

Request a TTS test route See the specification Bulk capacity available

At a glance HTTPS REST API or SIP interconnect Plain text, SSML, or a pre-recorded audio file Multiple languages and regional accents Optional DTMF capture for confirm / opt-out

Specification

The numbers, up front

Destination-level coverage

Connection: HTTPS REST API or SIP interconnect
Input: Plain text, SSML, or a pre-recorded audio file
Voices: Multiple languages and regional accents
Interaction: Optional DTMF capture for confirm / opt-out
Concurrency: Provisioned per account and per campaign window
Reporting: Per-call disposition, duration and DTMF response in CDRs
Best for: Mass notification, reminders and IVR prompts
Typical volume: Agreed per account, burst capacity on request

What you get

What TTS routing has to get right

Bulk voice is easy to sell and hard to deliver. The constraints that decide whether a campaign completes are concurrency, answer rate and audio clarity, in that order.

Concurrency sized to the window

A hundred thousand calls that have to complete inside an hour is a concurrency problem, not a volume problem. We provision channels against your deadline before the campaign starts.

Audio people actually understand

Natural-sounding synthesis with regional accents, or your own recorded prompt with dynamic fields inserted. A robotic voice gets hung up on.

Answer-rate aware routing

Time-of-day and per-destination answer behaviour feed route selection, because a call placed at the wrong hour is money spent on a voicemail.

DTMF capture where you need a response

Press 1 to confirm, 9 to opt out. Responses come back in the CDR so your system can act on them without a second channel.

Retry logic you control

Configurable re-attempts on no-answer and busy, with a cap, so a campaign does not become harassment.

Full disposition reporting

Answered, no answer, busy, failed, plus duration and DTMF, exportable per campaign.

The three things that decide whether a voice campaign works

Concurrency. Total volume is rarely the constraint. The constraint is how many calls can be in progress simultaneously, because that is what determines whether a hundred thousand reminders finish inside the hour they were useful in.

Answer rate. A call placed at the wrong hour, from the wrong CLI, into a market that treats unknown numbers as spam, is money spent on a voicemail. Route and pacing decisions have to account for local answering behaviour.

Audio. Synthesised speech that sounds wrong gets hung up on in the first two seconds, which is before the message has been delivered. Regional accent and natural pacing matter more than they sound like they should.

Why synthesise once

The audio for a campaign is generated once and reused across every call, with only the personalised fields regenerated. That is the difference between a broadcast that scales and one that becomes a synthesis bottleneck at the exact moment it needs to be fast.

Regulation is local

Permitted calling hours, mandatory opt-out mechanisms and CLI requirements differ by market and are enforced unevenly. We apply the local rule per destination rather than a single global policy, and we will tell you when a campaign as designed is not permitted somewhere on your list.

Use cases

What operators broadcast with TTS

Payment and collection reminders

Due-date and overdue reminders for lenders, utilities and subscription businesses, in the customer's own language.

Emergency and mass notification

Weather, outage and public-safety alerts where reach in the first ten minutes is the whole point.

Appointment reminders

Clinics, service businesses and logistics windows, with DTMF confirm to cut no-shows.

IVR prompts and announcements

Dynamic prompts generated from text rather than re-recorded every time a detail changes.

Service and outage notification

Telling customers something is broken before they call the contact centre to tell you.

Opted-in campaign broadcast

Promotional voice broadcast where local regulation and your own opt-in records permit it.

How it works

How a TTS campaign runs

Submit the text and the list

One API call per number, or a bulk job. Personalisation fields are inserted at synthesis time.
We synthesise once and fan out

Audio is generated once and reused across the campaign, which is what makes high concurrency affordable.
Calls are placed on answer-aware routes

Route and pacing are chosen against the destination and the completion window you set.
Dispositions return in real time

Per-call outcome and any DTMF response, so retries and follow-ups can start before the campaign finishes.

Connect

How to connect TTS Voice

Every method reaches the same routes, the same balance and the same reporting. Pick the one that suits how your platform is built.

Compare all connection methods

SMPP v3.4 bind

SMPP v3.4

An SMPP v3.4 bind is the standard way to move A2P SMS at volume. Transmitter, receiver or transceiver, with real…

Set up SMPP v3.4 bind

HTTP REST API

HTTPS / JSON

A straightforward HTTP REST API in JSON, for teams that would rather not run a persistent SMPP session. Same routes,…

Set up HTTP REST API

SIP interconnect

SIP / RTP

A SIP interconnect for voice termination, Voice OTP and TTS, with an engineer on the call for interop.

Set up SIP interconnect

IP authentication

IP allow-list

IP authentication instead of credentials: authenticate by source address, for platforms with fixed egress IPs.

Set up IP authentication

Coverage

Destination availability

Search for the destinations you care about. Anything not listed is usually available on request, capacity is added weekly.

Route availability by destination
Destination	Dial / MCC	A2P SMS	Voice OTP	TTS voice	Termination
Bangladesh	+880 · 470	Live	Live	Live	Live
India	+91 · 404 / 405	Live	Live	On request	On request
Pakistan	+92 · 410	Live	Live	On request	Live
Nepal	+977 · 429	On request	Live	On request	On request
Sri Lanka	+94 · 413	On request	Live	On request	On request
Indonesia	+62 · 510	Live	On request	On request	On request
Philippines	+63 · 515	Live	On request	On request	On request
Vietnam	+84 · 452	On request	On request	No	On request
Malaysia	+60 · 502	Live	On request	On request	On request
United Arab Emirates	+971 · 424	Live	On request	No	On request
Saudi Arabia	+966 · 420	Live	On request	No	On request
Oman	+968 · 422	On request	On request	No	On request
Qatar	+974 · 427	On request	On request	No	On request
Egypt	+20 · 602	On request	On request	On request	On request
Nigeria	+234 · 621	Live	On request	Live	On request
Kenya	+254 · 639	On request	On request	On request	On request
South Africa	+27 · 655	On request	On request	No	On request
Turkey	+90 · 286	On request	On request	No	On request
United Kingdom	+44 · 234	Live	On request	No	On request
United States	+1 · 310	On request	On request	No	On request

See 14 more destinations Show fewer

Why SisBird

Why TTS campaigns fail elsewhere

Most TTS problems are capacity problems

A campaign that trickles out over six hours instead of one has failed even if every call eventually connected. We size channels against the deadline, not the total.

Compliance is destination-specific

Automated voice is regulated differently market by market, permitted hours, mandatory opt-out, CLI rules. We apply the local rule rather than a global default.

The same account runs your OTP traffic

Notification and verification voice share credentials, balance and reporting.

Who runs these interconnects

Next step

Run a TTS broadcast on a real route before you plan a campaign

Tell us the destination, the concurrency you need and the window it has to complete in. We size the capacity and give you a test account to prove it.

1
You send the brief Destinations, monthly volume and the traffic type. Two lines is enough.
2
A routing engineer replies A costed plan with the route type stated per destination, usually within one business day.
3
You test on funded credit Same interconnects production runs on. No minimum term and no setup fee.

Or reach the routing team directly

Request a test route

For TTS Voice. We reply within one business day.

Reply within one business day Free test credits on approval No spam, no reselling of your data

FAQ

TTS voice routing questions

What is a TTS voice route?

A TTS (text-to-speech) voice route places automated calls in which synthesised speech reads out a message. It is used for reminders, alerts, IVR prompts and opted-in broadcast, and is billed like any other voice traffic.

How many calls can run at once?

Concurrency is provisioned per account. The right question is not how many calls in total but how many need to complete inside your window, tell us the deadline and we size the channels to it.

Can the call capture a keypress?

Yes. DTMF capture is supported for confirmations, opt-outs and simple menu choices, and the response is returned in the call record.

Can I use my own recorded voice?

Yes. Upload a recording and we insert dynamic fields such as names, amounts or dates into it, or use synthesis for the whole message.

Is CLI supported on TTS calls?

Where the destination market allows it. Some regulators require a registered CLI on automated calls, some prohibit foreign CLI entirely. The rule is recorded per destination in our coverage table.

Is promotional voice broadcast allowed?

Only where local regulation and your own opt-in records permit it, and only within permitted calling hours. We will decline traffic that puts your CLI or our interconnects at risk.

Read every wholesale routing question we are asked

Also available