Dograh Voice AI Tutorial - Open-Source Alternative to Vapi | OSS voice agents
Dograh is an open source alternative to Vapi. It is an OSS voice AI agent conversation builder.
In this video, we walk through Dograh, an open-source platform for building AI voice agents and voice automation workflows.
Dograh is designed as a developer-friendly alternative to platforms like Vapi, Bland, and Retell, allowing teams to build and deploy voice AI agents with full control over models, telephony, workflows, and integrations.
This walkthrough shows how to:
• Create a voice agent from scratch
• Build conversational workflows using nodes
• Extract structured data from conversations
• Connect APIs and external tools
• Add knowledge bases and documents
• Trigger actions through tool calls
• Integrate telephony for inbound and outbound calls
• Transfer calls to humans
• Analyze conversations using QA and analytics
Dograh supports multilingual voice agents, custom models, telephony integrations, API automation, knowledge bases, and advanced debugging tools that help developers understand exactly what happens inside an AI voice call.
If you're building AI phone agents, voice assistants, customer support automation, or outbound calling systems, this video gives a complete overview of how the platform works.
Resources
Docs
https://docs.dograh.com
GitHub
https://github.com/dograh-hq/dograh
YouTube
/ @dograhai
If you like the project, consider starring the repository on GitHub and following the channel for more tutorials.
#aicallagent #openvoiceai #opensourceai #openvoiceai #aicallagents #dograhai #dograhdemo #dograh #voiceaiopensource #opensourceconversationalai #opensourceaicallcenter #opensourcecallagents #openSourceVapiAlternative #OSS #opensourcevoiceai #opensourcevoiceagent #aicallagents #openvoiceai #opensourceai
00:00 - About Dograh
00:16 - Creating and Testing Your First Voice Agent
01:13 - Viewing Call Transcripts and Debug Data
02:05 - Debugging Voice Agents with Traces
03:00 - Understanding the Voice Workflow Editor and Nodes
05:30 - Extracting Structured Variables from Conversations
06:58 - Adding Knowledge Base Documents
07:50 - Creating Tools and API Integrations
10:00 - End Call Logic and Call Completion Scenarios
12:00 - BYOK and Choosing LLMs, Voices, and Transcription Models
13:27 - Transferring Calls to Human Agents
14:27 - Deployment & Custom Dictionary
15:00 - Voicemail and Post call QA Analysis