What Is The Difference Between Dograh's Cascade And Speech-To-Speech Pipelines?

The cascade pipeline routes audio through three sequential hops — STT → LLM → TTS — which adds latency but gives granular control over each model. Speech-to-speech sends raw audio directly into a model like Gemini Live or GPT Realtime and receives audio back, eliminating the text round-trip for ultra-low latency and more natural turn-taking.

Can Dograh Handle Both Inbound And Outbound Calling On The Same Platform?

Yes. Dograh supports inbound and outbound workflows natively through its telephony integrations with Twilio, Vonage, Telnyx, Plivo, Cloudonix, and Asterisk SIP. Outbound campaigns include voicemail detection, redial scheduling, and bulk contact list management from the same dashboard.

How Does Dograh's MCP Server Integration Work With Developer IDEs?

Dograh ships a Model Context Protocol (MCP) server that exposes the platform's workflow API to MCP-compatible clients like Claude Code, Cursor, or Codex. Developers can spin up, modify, and deploy full voice agents directly from the terminal or IDE without opening the visual builder.

What Self-Hosted STT And TTS Models Does Dograh Support Out Of The Box?

Dograh's self-hosted stack currently supports Whisper and Voxtral for STT, and Kokoro and Chatterbox for TTS, all running inside your own VPC. LLM self-hosting is in beta. You can also bring custom voice clone weights to the TTS layer.

Does Dograh Provide LLM Call Tracing And Observability Without A Third-Party Tool?

Yes. Dograh has a built-in LLM observability and analytics dashboard alongside a dedicated Traces view for analyzing LLM call behavior in real time. The QA node in the workflow builder further lets you score and evaluate prompt performance across individual nodes without leaving the platform.

Dograh

Name: Dograh
Availability: InStock
Author: Zansat Technologies Private Limited (Dograh)

by Zansat Technologies Private Limited (Dograh) • Bengaluru, Karnataka, India • Founded 2024

Open-Source Voice Agent Platform Built For Production-Grade AI Calling

No reviews yet

3 1

voice-ai open-source voice-agents self-hosted ai-telephony

Visit Dograh

Follow:

Trust Score

Based on ratings & reviews

4.6 /5

49 reviews

Views

Clicks

Pricing

Free forever (self-hosted OSS), Pro on custom pricing (managed cloud)

What is Dograh?

Dograh is a fully open-source, self-hostable voice AI agent platform — the developer-first alternative to Vapi and Retell. Built on a Pipecat-based engine under a BSD 2-Clause license, it lets teams design inbound and outbound call workflows using a no-code drag-and-drop builder, then deploy on their own infrastructure or Dograh's managed cloud.

Bring your own STT, LLM, and TTS keys — or use Dograh's default stack. With built-in telephony integrations (Twilio, Vonage, Telnyx, Plivo), MCP server support, real-time LLM observability, and a hybrid human-voice-plus-TTS engine, Dograh eliminates vendor lock-in and platform fees entirely.

Dograh — Open-Source Voice Agent Platform Built For Production-Grade AI Calling Whether you're evaluating Dograh for your team or comparing it to alternatives in the AI Code Tools category, this in-depth review covers everything: features, pricing, real user reviews, pros and cons, integrations, and direct comparisons against competitors.

Key Features 8

Hybrid Pre-Recorded + TTS Engine Cuts Cost Up To 3× And Boosts Conversion

Speech-To-Speech Audio Pipeline With Real-Time Turn-Taking And Interruption Handling

MCP Server Lets IDEs Like Cursor Or Claude Code Deploy Voice Agents Programmatically

Built-In QA Node Analyzes Prompt Quality Across Workflow Nodes Automatically

Modular BYOK Architecture Supports Any STT, LLM, Or TTS Provider Or Self-Hosted Model

Multi-Telephony Config Support Across Twilio, Vonage, Telnyx, Plivo, And SIP/Asterisk

Agent Versioning With Per-Node Model Configuration Override At Runtime

Outbound Campaign Manager With Voicemail Detection And Redial Options

Who Is Dograh For

1 Backend Developers

2 DevOps Engineers

3 AI Startup Founders

4 SaaS Builders

5 Enterprise Voice Automation Teams

Pros & Cons

Pros 4 benefits

Zero Platform Or Per-Minute Fees
Full Source-Level Customization
Weekly Production-Grade Releases
YC Alumni-Backed Open Source

Cons 3 limitations

No Native GUI Mobile App
Pro Pricing Lacks Transparency
Self-Hosting Requires Docker Knowledge

Frequently Asked Questions

5 questions

How Dograh works

Dograh is positioned as open-Source Voice Agent Platform Built For Production-Grade AI Calling. Under the hood it ships 8 headline capabilities, including Hybrid Pre-Recorded + TTS Engine Cuts Cost Up To 3× And Boosts Conversion, Speech-To-Speech Audio Pipeline With Real-Time Turn-Taking And Interruption Handling, MCP Server Lets IDEs Like Cursor Or Claude Code Deploy Voice Agents Programmatically, Built-In QA Node Analyzes Prompt Quality Across Workflow Nodes Automatically, Modular BYOK Architecture Supports Any STT, LLM, Or TTS Provider Or Self-Hosted Model and Multi-Telephony Config Support Across Twilio, Vonage, Telnyx, Plivo, And SIP/Asterisk. Together these features cover the core workflows most teams expect from a modern ai code tools, from initial setup through day-to-day production use.

Dograh runs as a self-contained product, so you can adopt it without touching the rest of your stack — useful when you want to evaluate the tool in isolation before wiring up integrations.

Who is Dograh for?

Dograh is most useful for Backend Developers, DevOps Engineers, AI Startup Founders and SaaS Builders. If your team falls into one of those buckets, the feature set lines up well with how you already work — you won't be forcing a square peg into a round hole.

Beyond the obvious use case, the product tends to attract users who want a low-friction starting point option in the ai code tools space.

Dograh pricing explained

Dograh runs on a freemium model. You get a usable free tier to evaluate the product, and you only pay when you outgrow the limits — usage volume, seat count, or premium features. Headline pricing: Free forever (self-hosted OSS), Pro on custom pricing (managed cloud).

Across the AI Gear Base rubric, we score freemium pricing models on transparency, rate-limit honesty, and how predictable spend is at scale. Dograh's freemium approach is standard for the category — useful for evaluation, but always re-check tier limits before you depend on the free plan.

Our verdict on Dograh

Dograh hasn't been rated by enough reviewers yet to publish an aggregate score. The strongest signal in those reviews is that zero platform or per-minute fees. The most common complaint is that no native gui mobile app — worth knowing before you commit, but rarely a deal-breaker for teams that already match the use case.

If you're evaluating Dograh against alternatives, weigh it on the same 7-criteria rubric we apply to every tool: capability, integrations, pricing transparency, support, security posture, roadmap velocity, and community signal. Built by Zansat Technologies Private Limited (Dograh), founded in 2024, the product has a clear track record you can verify before adopting it. The bottom line: Dograh is a solid pick in the ai code tools category, and it deserves a spot on your shortlist if your workflow matches what it was built for.

What's New

weekly

Agent Skills, Chat-Based Testing, And MCP Search Docs 1.31.0

Added agent skills for PR review, chat-based testing mode for voice agents, Tuner integration for observability, and a new MCP search_docs tool over the full documentation corpus.

May 21

OpenAI Realtime Model Support 1.30.0

Added support for OpenAI Realtime models inside the speech-to-speech pipeline, enabling ultra-low-latency audio-in/audio-out conversations through GPT Realtime 2.

May 16

View all updates

User Base

1M+ minutes of AI calls processed on cloud platform

Active Users

Security & Privacy

Self-hosted (any region), Managed cloud, Private VPC

Full Data Sovereignty Via Self-Hosted Deployment Open-Source Codebase — Fully Auditable By Internal Security Teams BYOK (Bring Your Own Keys) — No Vendor Access To API Credentials No Vendor HIPAA, GDPR, Or SOC 2 Dependency When Self-Hosted Data Residency Options Available Via Private Cloud Managed Deployment