ClawdTalk provides a voice interface layer that connects existing text-based bots to the public telephone network. It handles the entire voice pipeline, including speech-to-text, text-to-speech, and call management, allowing users to interact with their bots via natural voice commands. This service eliminates the need for developers to manage complex telephony infrastructure or expose their bots publicly via gateways.
Funding
$47M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
MFounders
Product
Problem
Developers and businesses that build conversational AI bots must manage complex telephony stacks—provisioning phone numbers, handling audio streams, performing speech‑to‑text and text‑to‑speech, and exposing services to the public internet—to enable voice interactions. This overhead limits rapid deployment and raises security and latency concerns.
Solution
ClawdTalk abstracts the entire voice‑call workflow into a hosted service. Bots receive plain text events and return text responses; ClawdTalk handles inbound and outbound PSTN calls, real‑time transcription, and natural‑sounding synthesis using Telnyx’s AI Assistants. Calls are routed through a global, carrier‑grade network with sub‑200 ms conversational latency, while a persistent outbound WebSocket keeps the bot behind NATs and firewalls, eliminating the need for public endpoints. Optional PIN protection and dedicated numbers add access control for shared or production environments. The platform exposes simple REST endpoints for call initiation and delivers structured JSON payloads, enabling developers to add voice channels without telephony expertise.
Target Audience
The primary customers are AI bot developers, SaaS platforms, and enterprises that need to add voice interaction to existing text‑based conversational agents without building telephony infrastructure.
Features
- Two‑way PSTN calling: inbound calls to a bot or outbound calls initiated by the bot via a single `POST /v1/calls` API.
- Real‑time speech‑to‑text and text‑to‑speech powered by Telnyx NaturalHD models (latency < 200 ms, HD Voice codec AMR‑WB).
- Persistent outbound WebSocket connection for private, firewall‑friendly bot integration; no port‑forwarding or public IP required.
- PIN‑based call authentication with bcrypt‑hashed storage, supporting multiple attempts and per‑number access rules.
- Global carrier‑grade network with private fiber and direct carrier links (latency < 100 ms, 99.999 % uptime).
- Structured JSON event format for inbound messages and response payloads, simplifying bot logic.
- Dedicated phone numbers and outbound call capability on paid tiers, enabling full conversational workflows.
- Built‑in analytics: call minutes, text usage, and optional call recordings (future feature).