Request validation and parsing for incoming Twilio webhooks.

Implements the OpenAI Realtime API for voice conversation with audio streaming.

Resamples 16-bit PCM audio to 8kHz telephony sample rate using linear interpolation.

Orchestrates bidirectional audio streaming between telephony and AI providers.

OpenAI-compatible TTS & STT API — Phase 3.2 (ROADMAP.md P0). Drop-in replacement for OpenAI's audio endpoints so that any tool speaking the OpenAI protocol (Claude, Cursor, LangCha

Voice layer: the VoiceProvider Protocol, shared IO models, and all three run-mode providers (mock, replay, live Vapi). mock/replay read canned transcripts from samples/transcripts/

Indian-phone normalization. Shopify checkout returns phones in many shapes: +91 98765 43210 (formatted E.164) 919876543210 (no plus, with country code) 09876543210 (leading zero, n

Immutable transcript segment with start time, end time, and text content.

React hook for voice recording using MediaRecorder API. Records audio in WebM/Opus format for efficient streaming. /

Initiates outbound calls through a telephony provider, handling state management, configuring duration and transcript timers, and persisting call records.

Validates Twilio webhook signatures using HMAC-SHA1 with replay attack prevention through time-windowed request caching.

Captures and snapshots Android UI hierarchy from interactive windows for OpenPhone assistant, providing accessibility event monitoring and node retrieval.

Do-Not-Disturb window enforcement for Indian telecom compliance. TRAI (Telecom Regulatory Authority of India) restricts commercial voice calls to 09:00–21:00 IST. Calling outside t

The canonical status vocabulary. call-api owns it; everything else is a dialect. The voice agent has its own words for how a call ended (`goal_achieved`, `no_audio`, `abandoned`, `

Pydantic model validating call initiation requests with a phone number field, extending a workflow initiate model for call orchestration.

zvonok voice agent — one goal-directed outbound PSTN call per job. Dispatched with metadata: {"number": "+34...", "goal": "...", "language": "en", "caller_id": "+44...", "max_durat

Ten protocol adapters (OpenAI Realtime, Deepgram Voice Agent, Twilio Media Streams, ElevenLabs, Gemini Live, Pipecat, Vapi, AssemblyAI, Cartesia, Native) handling wire-format negot

Checks whether a SIP response code is one of the known invalid number codes (404, 410, 484, 604) for carrier failure handling.

State machine managing lifecycle transitions for active phone calls.

Processor class for VibeVoice ASR models.

React hook for handling SSE voice streaming responses. Converts audio blob to base64 and sends as JSON to match server expectations. /

Abstract speech recognition interface providing flexible model integration and transcription methods.

Parses raw phone input, converts to standard E.164 format for telephony.

Everything the agent says, and everything that checks it said it. Two kinds of text live here and they are not the same kind of thing: FIXED TEXT — the disclosure and the probe. Sp

AVAudioEngine-based audio recorder with PCM buffer tap, audio session management, interruption handling, and level metering. Lifts out as a standalone audio capture utility.

LiveKit audio observer. Transcribes a participant's actually-published audio track into an ObserverTranscript via STT. record_speech_events is the architecture- and transport-indep

Generates Asterisk configuration for realtime media, defining RTP/WebSocket transport types and related settings.

In-process Voicelink mock that satisfies TelephonyProvider. Used to develop and test the rest of the platform (campaign engine, call lifecycle, billing) before Voicelink Q1/Q2/Q3 a

Orchestrates VAD→STT→LLM→TTS pipeline with barge-in, silence detection, call recording, and ntfy push notifications. The core brain of a real-time voice AI.

TTS provider registry

Provides a provider-agnostic protocol and events for real-time speech-to-speech voice sessions, supporting multiple backends including Qwen, Gemini, and OpenAI Realtime.

Reusable AudioWorklet for streaming PCM16 audio playback. Place in public/ folder and load via audioContext.audioWorklet.addModule() /

Resolves and returns validated voice call configuration settings from environment or runtime sources.

Calculates pricing estimates for voice call stacks by summing platform, carrier, recording, STT, LLM, and TTS costs based on scenario configuration.

https://github.com/mohayonao/wav-decoder

Provides Twilio voice call integration handling outbound/inbound calls, webhook verification, TwiML responses, and call status normalization for telephony applications.

Records user audio via mic, encodes to webm/blob, handles permissions, uploads backend.

Serializes a CallLog into a JSON-ready dict, converting enums, timestamps, and transcript entries while guarding optional timings to avoid crashes during status broadcasts.

MiniMax text-to-speech SDK wrapper.

Kotlin call state graph managing call lifecycle transitions and signaling events.

Application Configuration Manages all environment variables and settings using Pydantic Settings.

Swappable STT / LLM / TTS factories driven by config. Every provider is constructed here so the rest of the pipeline never imports a specific vendor. Switch providers by changing L

Main voice agent bot with call handling, transcription, and LLM integration.

Wraps OVH API callers per json impl.

Pure C G.711 mu-law/a-law encoder/decoder and L16 linear PCM utilities with canonical ITU algorithms. No external dependencies, fully unit-tested off-platform.

Advanced HTML5 audio engine with HLS.js support, Media Source Extensions for fMP4 streaming, crossfade, equalizer, oscilloscope visualization, gapless playback, and stall detection

Audio utility functions for voice chat. Handles PCM16 decoding and AudioContext initialization. / * Decode base64 PCM16 audio to Float32Array for Web Audio API /

Defines audio client configuration settings for telephony conversations (codec, jitter buffer, silence detection) within IQRA.

Initiates Twilio voice calls, returns call/summary responses, and handles API errors via typed client.

Processes incoming call events, enforcing inbound policies (allowlist, open, disabled) and coordinating call lifecycle transitions including acceptance, rejection, and state persis