Build a voice AI agent that answers calls, qualifies leads, books appointments, and transfers to a human when needed, all from a drag-and-drop workflow builder in your browser. Dograh ships as a Docker Compose stack (API, web UI, Postgres, Redis, MinIO) that you self-host on any Linux server with automatic HTTPS provisioning via Let's Encrypt. The visual workflow builder lets you design multi-turn conversation flows by connecting nodes for greetings, intent classification, tool calls, and handoffs; describe your use case in plain English and the platform generates the LLM prompt and node graph for you. Connect your own speech-to-text, LLM, and text-to-speech providers (OpenAI, Anthropic, Gemini, ElevenLabs, Deepgram, local Whisper, Kokoro, or any OpenAI-compatible endpoint) or use the built-in Speech-to-Speech mode with Gemini Flash Live and GPT-Realtime-2 for sub-200ms latency. Telephony plugs in through Twilio, Vonage, Vobiz, or Cloudonix for inbound and outbound calling, with live agent transfer when the conversation needs a human. Webhook tool calls connect to Salesforce, HubSpot, Google Calendar, Cal.com, or any REST API without writing orchestration code. The ClonedVoice feature mixes real human voice recordings for high-frequency phrases with neural TTS fallback for dynamic content, cutting costs while improving caller trust. A built-in MCP server lets AI coding agents like Claude Code or Cursor design, test, and edit workflows through natural language. Deploy on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSD 2-Clause licensed.

Your front desk phone rings at midnight and nobody sane is awake; QuickVoice answers it with an AI agent you configured, on servers you own, with call recordings you control. Build agents in the web console: write the prompt, pick speech-to-text, language model, and voice, set the first message, then link a real phone number from Twilio or Telnyx through LiveKit SIP trunks. Inbound callers get a receptionist that answers questions from your uploaded knowledge base, PDFs and web pages chunked and embedded for live retrieval, while outbound mode fires single quick calls or batch campaigns for reminders and follow-ups. Agents can invoke allowlisted MCP tools mid-call, so checking an order status or booking a slot happens inside the conversation instead of after it. Every call lands in the log with transcript, recording, and metadata, searchable from the analytics dashboard, and privacy controls redact personal data by default, can suppress recordings entirely, and sweep old transcripts on a retention schedule. Multi-language voices, per-agent runtime overrides for campaigns, organization accounts with API keys and audit logs, and Stripe billing hooks round out a platform built for teams who would rather inspect their voice stack than rent a black box. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Once a voice agent is imported or built, Lampline connects it to a phone line, sends an AI caller to grade real calls, and keeps it all on your own server. Start by importing a Dialogflow CX or CX Agent Studio export along with a report of anything to check, or build fresh in a drag-and-drop builder that mixes LLM steps with scripted pages for exact lines and form fields. Next, pick one of 20 model providers, including Claude, OpenAI, Gemini, Bedrock, Groq, and DeepSeek, or run Ollama with local speech and no API keys. Test by chat or microphone while each step lights up, and let agents call HTTP and MCP tools and save records in collections of their own. Then connect callers through Twilio, Telnyx, or any SIP trunk, PBX, or session border controller via jambonz, with transfers to a person that carry the call's details as SIP headers. Finally, evals phone the agent over a real line, where an AI Test Caller brings background noise, accents, and interruptions, grades each transcript check by check, and reports typical and p95 response times as callers hear them. Each agent stays a folder of YAML and Markdown that you can review in git and lint or test with the vk CLI. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

MikoPBX gives a small business a complete phone system on a server it controls, with extensions, call queues, IVR menus, and call recording instead of per-seat hosted PBX fees. Follow a call through it: incoming rules per provider match exact numbers or patterns, fall through by priority when nobody answers, and end at a default route. Each employee gets an extension with SIP credentials, forwarding for no-answer, busy, and unavailable states, and an optional recording switch. Queues offer seven distribution strategies, including Ring All, Least Recent, Fewest Calls, and Memory Hunt, and can announce each caller's position and estimated wait. IVR menus play a recorded greeting, route keypad choices to departments or extensions, and fall back to a default number after a set number of retries. Call history filters by caller, destination, or DID, flags missed calls in red, and plays or downloads recordings as WebM, MP3, WAV, or OGG. Voicemail, call transfer, call pickup, and a holiday and night-time switcher serve daily needs, while free and paid marketplace modules, among them a softphone, auto dialer, and LDAP/AD sync, handle the rest. A firewall, Fail2Ban, passkey sign-in, and CrowdSec integration protect the system, and a REST API with more than 250 endpoints lets other tools manage it. The interface speaks 26 languages. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPL-3.0 licensed.

FreeSBC sits at the edge of your phone system so your FreeSWITCH or Asterisk never has to face the public internet. It is the only element with a public address: phones register through it over SIP on UDP, TCP, or TLS, browsers call through WebSocket with WebRTC terminated at the edge (ICE-Lite, DTLS-SRTP, and RTCP-mux, relayed onward as plain RTP), and carriers reach your switch through it as an outbound proxy, with the switch's private addresses hidden from every SDP body, Contact, and Request-URI it constructs. The switch stays on a private LAN as the registrar and call brain, holding credentials and carrier accounts, while FreeSBC anchors all media on its own ports with hardened first-packet latching and a silence watchdog, and never transcodes. Against the background radiation of SIP scanning, it applies per-IP rate limits with separate limits for carrier sources, scanner fingerprinting with in-memory bans, and admission control for public INVITEs and REGISTER enumeration, and every denial is a silent drop. A multi-switch pool hashes users across switches, and one fixed private socket keeps the trusted side simple. The embedded WebUI, REST API, and Prometheus metrics sit behind a bcrypt-checked admin login, showing live calls and media ports. One static binary, one YAML file, no database, no Redis, and no separate media process to babysit. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

With over 20,000 GitHub stars and adoption by companies building everything from telehealth platforms to AI voice agents, LiveKit is the most widely deployed open-source real-time communication server available. The Go-based Selective Forwarding Unit handles hundreds of concurrent participants per node with adaptive bitrate streaming, simulcast layers, SVC codec support for VP9 and AV1, and end-to-end encryption. Client SDKs span JavaScript, Swift, Kotlin, Flutter, React Native, Rust, Python, Unity, and ESP32 embedded devices, while server-side APIs cover Node.js, Go, Ruby, Java, Python, Rust, PHP, and .NET. The Agents framework enables building AI-powered voice and video applications — real-time speech-to-text, LLM-driven conversations, and computer vision pipelines — running as server-side participants in any room. Egress records sessions to S3-compatible storage or streams to RTMP endpoints, while Ingress pulls external feeds from OBS via RTMP, WHIP, or SRT into LiveKit rooms. The SIP bridge connects traditional telephony to WebRTC rooms for hybrid conferencing. JWT-based authentication, webhook notifications, room-level moderation APIs, and selective subscription give operators granular control. Deploy as a single binary for development, Docker Compose for production single-node, or Kubernetes with the official Helm chart for distributed multi-region clusters using Redis for state coordination. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

One Janus installation can host a video meeting in one session, stream a camera feed in another, and bridge a browser call into a telephone network in a third, because Janus is a general-purpose WebRTC server whose work is done by interchangeable plugins. The VideoRoom plugin publishes each participant's streams to everyone else in the room with simulcast and viewport selection, so a class or team call adapts quality per viewer instead of forcing one bitrate on all. Streaming mounts live or prerecorded sources, including RTSP cameras and RTP feeds, and hands them to any number of WebRTC viewers, while AudioBridge mixes voices into conference rooms with recording and signed-token access. SIP and NoSIP plugins register with existing VoIP infrastructure and place or receive calls, translating between WebRTC media and telephone audio so a web page can ring a desk phone. Record&Play captures publisher streams to files you can replay through the same server, TextRoom adds data-channel chat rooms, and Lua or JavaScript plugins let you script custom media logic without recompiling the core. Applications talk to Janus over HTTP or WebSocket APIs, with RabbitMQ, MQTT, and NATS transports for bus-driven deployments, and event handlers report session activity to your monitoring stack. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPL-3.0 licensed.

With 26,000 GitHub stars and consistent victories over ElevenLabs in blind evaluations, Chatterbox delivers state-of-the-art text-to-speech with zero-shot voice cloning requiring only 5 seconds of reference audio. The model family spans three architectures: Chatterbox Multilingual V3 (500M parameters, 23+ languages including Arabic, Chinese, Japanese, Korean, Hindi, French, German, Spanish, and Portuguese), Chatterbox-Turbo (350M parameters optimized for voice agents with a single-step distilled decoder achieving ~200ms time-to-first-speech), and Chatterbox-Nano (110M parameters running 3x faster than realtime on 8 CPU cores for edge deployment). Unique among open-source TTS systems, Chatterbox introduces emotion exaggeration control — adjusting intensity from monotone to dramatically expressive via a single parameter — and native paralinguistic tagging where tokens like [laugh], [cough], [chuckle], and [gasp] inject natural vocal reactions inline without post-processing. The alignment-informed inference pipeline eliminates hallucinations and repetition artifacts common in autoregressive TTS. Built-in PerTh neural watermarking embeds imperceptible forensic identifiers in generated audio for provenance tracking. Trained on 500,000 hours of cleaned speech data across all supported languages. Voice conversion scripts enable transforming existing audio into any cloned voice. Deploy via pip install with PyTorch, serve through Gradio interfaces or custom FastAPI endpoints, and expose via HTTP streaming or WebSocket for sub-200ms conversational applications. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Kokoro-FastAPI turns text into natural-sounding speech across eight languages by serving the 82-million-parameter Kokoro-82M model through an OpenAI-compatible REST API, so any existing OpenAI SDK client can generate audio by just changing the base URL. With over 5,300 GitHub stars since December 2024, the fully Dockerized FastAPI server covers American English, British English, Spanish, French, Hindi, Italian, Japanese, Brazilian Portuguese, and Mandarin Chinese with language-specific phoneme processing. Inline voice mixing blends multiple profiles using weighted ratios like af_bella(2)+af_heart(1), automatically normalizing weights and caching combined voicepacks as PyTorch tensor files for reuse. Audio streams in real time over HTTP with configurable chunk sizes, or generates complete files in MP3, WAV, OPUS, FLAC, AAC, or PCM formats with speed control from 0.25x to 4.0x. Per-word timestamped captions with speaker-tagged voice labels enable subtitle generation for podcasts, audiobooks, and accessibility workflows. Pre-built Docker images support NVIDIA GPU acceleration via CUDA, experimental AMD GPU inference via ROCm, and CPU-only deployment on linux/amd64 and linux/arm64 architectures, with Apple Silicon MPS support available through direct UV execution. The integrated web interface at port 8880 provides browser-based speech generation, while the Swagger UI at /docs exposes the full API reference. Debug endpoints report system statistics for monitoring inference load. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

With over 48,000 GitHub stars and monthly releases since March 2023, LocalAI is the self-hosted AI engine that replaces every OpenAI endpoint with a single Docker container running on your own infrastructure — serving chat completions, image generation, text-to-speech, speech-to-text, embeddings, vision, video generation, and function calling through identical API schemas that require zero application code changes. The composable backend architecture isolates each inference engine as a separate gRPC service running in its own OCI container, so llama.cpp, vLLM, SGLang, transformers, whisper.cpp, diffusers, MLX, Stable Diffusion, and Flux install on demand without touching the core, can run on separate machines, and a fault in one never affects others. Hardware acceleration spans NVIDIA CUDA 12 and 13, AMD ROCm, Intel oneAPI/SYCL, Apple Silicon Metal, Vulkan, and NVIDIA Jetson L4T — or runs entirely on CPU without any GPU. Built-in AI agents support autonomous tool use, retrieval-augmented generation, Model Context Protocol integration, and skill-based workflows directly in the web interface. The model gallery provides curated YAML configuration files for hundreds of models that install with a single command, while P2P federated inference distributes model shards across multiple machines for running models larger than any single node's memory. Multi-user API key authentication with quotas and role-based access enables team deployments. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

A fair-source chatbot and conversational-form builder: Typebot assembles conversations in a visual graph editor. In a visual graph editor you chain blocks from four categories: bubbles display text, images, video, audio, and embeds; inputs collect data through text fields, email, phone, buttons, picture choices, date pickers, file uploads, and Stripe payments; logic blocks handle conditional branching, variables, URL redirects, A/B testing, and custom JavaScript; integration blocks call webhooks, OpenAI, Google Sheets, Google Analytics, Meta Pixel, Zapier, Make, and Chatwoot. Build once, deploy anywhere: custom domains, WhatsApp, or embedded in any site as a container, popup, or chat bubble through a fast native JS library with no iframe and no external dependencies - plus an HTTP API for executing bots programmatically from any language. Theming covers fonts, colors, roundness, and shadows with custom CSS and reusable templates, and results arrive in real time with drop-off and completion analytics plus CSV export. Two Next.js apps (builder and viewer) self-host via Docker under the Functional Source License, which converts to Apache 2.0 after two years.

Build, deploy, and monitor chatbots and LLM-powered agents on one open-source conversational AI platform: Botpress. Its Studio is a visual development environment: a drag-and-drop canvas arranges conversation logic with nodes for messages, questions, choices, and actions, while a built-in emulator simulates conversations for debugging before anything goes live. Agents ground their answers in a knowledge base assembled from uploaded documents, ingested websites, and past conversations via retrieval-augmented generation, and the LLM layer connects to multiple model providers - GPT-4, Claude, Mistral - with a configurable model strategy. An autonomous engine handles reasoning, tool orchestration, persistent memory across sessions, and sandboxed code execution, and custom code actions in TypeScript extend agents past prebuilt workflows. Over 100 integrations deploy the same bot to WhatsApp, Telegram, Slack, Microsoft Teams, and web chat, and connect it to HubSpot, Zendesk, Zapier, and arbitrary APIs and webhooks. Human handoff, conversation analytics, and quality monitoring cover production operation. Originating in 2017 from a Montreal team, the community edition is developed openly on GitHub.

Tiledesk lets you build AI-powered conversational agents with a visual drag-and-drop designer, then deploy them simultaneously across web chat, WhatsApp, Telegram, Facebook Messenger, Instagram, email, and Slack without rebuilding per channel. The no-code Design Studio combines LLM-powered conversations with conditional logic, external API calls, and branching dialogue trees that work identically everywhere. Multi-RAG knowledge management separates content into isolated repositories with hybrid search combining traditional full-text retrieval and semantic understanding via Qdrant vector embeddings, enabling accurate answers even when users phrase queries in unexpected ways. LLM integration supports GPT-4, GPT-4o, Anthropic Claude, and any model exposing an OpenAI-compatible API including locally hosted models via Ollama for complete data sovereignty. The human-in-the-loop system provides seamless escalation from AI agents to live support staff with full conversation context preserved, while multi-agent workflows enable complex orchestration where specialized bots collaborate on different aspects of a customer interaction. REST and MQTT APIs power integration with external systems and webhook-triggered automations, and pre-designed templates accelerate common scenarios. Docker Compose deployment starts the full stack including server, dashboard, messaging engine, and widget. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Call-Me puts a click-to-call video button on your website so a visitor can start an encrypted one-to-one video call with your team right in the browser, with no downloads, no accounts, and no phone numbers. A shared link carries everything: send a customer to a named room such as Support or Reception, or link straight to one person so the call rings them directly, which suits support desks, telehealth check-ins, tutoring, and sales consultations. Everyone in a room sees who else is present and available, and one click on a name starts the call. During the call you can chat in real time, share your screen, send files, switch between cameras, microphones, and speakers, hide your video, or mute yourself. Web push notifications flag an incoming call even when your tab sits in the background, so a staffed room never silently misses a visitor. An optional host password keeps the operator side of the room private. A REST API reports connected users, rooms, availability, and active calls, and can start a call for you, while webhooks post join, leave, and call start and end events to your other systems. The interface speaks multiple languages and adapts to phones and desktops alike. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

A visitor clicks your link and a face is already there, listening, answering within a fraction of a second, and stopping the moment they interrupt, because Mynah runs live digital humans in the browser instead of playing back rendered video. Underneath, cored takes WebRTC offers, runs the conversation through your chosen language model, speech recognition, and text to speech, and drives a lip synced avatar engine such as MuseTalk, FlashHead, or Wav2Lip, with a semantic end of turn model deciding when a person has finished speaking. The admin console is where you shape the experience: bake a new avatar from a talking video with explicit consent, clone or audition voices, swap avatars per channel without cold starts, and tune green screen keying and gestures live in the playground. Knowledge bases take your documents through ingestion and embedding so answers ground in your material, with retrieval tests before anything ships. Publishing freezes a model, voice, and knowledge snapshot into a channel: a shareable link with token, domain, or CIDR access control, its own branding, and a concurrency cap drawn from an engine pool that computes how many workers your VRAM can hold. Session monitoring shows who is connected and lets you kick a session, health checks watch every dependency, and light and dark themes with Chinese and English interfaces cover both console and visitor pages. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.