A voice agent that works for callers who are deaf or stutter
Texted at once, understood through a stutter, never rushed. Tested, with its limits stated.
Built for the business's own customers
People who are deaf, hard of hearing or who find speaking hard call small businesses too. A phone menu that only listens for a clear sentence loses them. The Speakvora agent has rules for this built in, tested with automated tests, and you can see and change the choice in your dashboard.
What it does
- Deaf caller or a request for a text: texted straight away; the conversation continues in writing.
- Stutter or hesitation: cleaned to what was meant; the record keeps what was actually heard.
- Slow or effortful speech: more time, "Take your time, I'm listening", then the parts are joined before it answers.
- Hard of hearing: slower, louder speech for the rest of the call and an offer of a text.
- Never told off: someone struggling to be heard gets time and a choice of text or person, not an error message.
Limits
It is not a medical or assistive device and it does not replace a relay service. Speech recognition comes from your phone provider, so unusual speech can still be misheard; that is why a text or a person is always one step away. It follows the rest of the agent's safety rules (consent, STOP, quiet hours).
You pay once per business location. Speakvora charges no licence fee per call, per text, per minute or per conversation turn; your phone provider bills your calls and texts, and an AI key is optional and billed by your AI provider. New lines the agent needs are voiced once through the Speakvora API at $3 per million characters, and a line you save to your library replays at no further charge; speech to text is $0.0015 a minute. Pricing · AI front desk for small business.
Questions
What happens when a caller is deaf or asks for a text?
The caller is texted at once and the conversation continues in writing. Businesses where being understood really matters (clinics, pharmacies, legal, finance, public services) are set to put a person first: the caller is texted, your team is told immediately and a person is connected.
Does it understand a stutter?
A stutter or hesitation such as "b-b-book a t-t-table, um, f-for four" is cleaned to what was meant ("book a table for four"), and the call record keeps both what was heard and what it was understood as. This works on the words the phone provider's speech recognition returns, so very heavy or unusual speech can still be misheard; a text or a person is always offered.
Does it rush slow speakers?
No. A caller who is still speaking or stops mid-sentence hears "Take your time, I'm listening", the agent waits longer and joins the parts together before answering.
What about callers who are hard of hearing?
A caller who says things like "speak up" or "I can't hear you" gets slower, louder speech for the rest of the call and an offer of a text.
Does this need an AI account or send audio anywhere?
No. These rules run on your own computer with no AI needed and nothing leaves it. You pay once per business location. Speakvora charges no licence fee per call, per text, per minute or per conversation turn; your phone provider bills your calls and texts, and an AI key is optional and billed by your AI provider. New lines the agent needs are voiced once through the Speakvora API at $3 per million characters, and a line you save to your library replays at no further charge; speech to text is $0.0015 a minute.
Can I choose how it behaves?
Yes. Three presets: offer a text first (restaurants and shops), a person first (clinics, pharmacies, legal, finance, public services) or a text or a ticket (app and product support). You can see and change the choice in Settings.