# Parlora v0.1.0 - Table of Contents > An Elixir framework for building real-time voice AI agents ## Pages - Getting started - [Parlora](readme.md) - [Your first working session](first-session.md) - [Getting started](using-parlora.md) - [Requirements and native builds](beam-compatibility.md) - [OrderDesk example](order-desk.md) - [OrderDesk workflows](order-desk-workflows.md) - Channels - [Browser WebRTC channel](browser-channel.md) - [Incoming Twilio calls](twilio-channel.md) - [Incoming Telnyx calls](telnyx-channel.md) - Providers - [OpenAI voice provider](openai-provider.md) - [Gemini voice provider](gemini-provider.md) - [Grok voice provider](grok-provider.md) - Backends - [Jido backend](jido-backend.md) - [Sagents backend](sagents-backend.md) - [Legion backend](legion-backend.md) - Reference - [Changelog](changelog.md) - [GPT Live output overload recovery](output-recovery.md) - [Adapter contracts and data reference](public-interfaces.md) - [MIT License](license.md) ## Modules - [Parlora.AdapterInit](Parlora.AdapterInit.md): Identity, event destination, agreed format, and limits supplied to an adapter. - [Parlora.AudioFormat](Parlora.AudioFormat.md): Explicit pipeline encoding and byte timing for PCM16LE and G.711 μ-law. - [Parlora.AudioFrame](Parlora.AudioFrame.md): An ordered audio fragment. Locally derived timestamps are explicitly labeled. - [Parlora.Backend.Capabilities](Parlora.Backend.Capabilities.md): Features supported by the configured backend mapping. - [Parlora.BackendRequest](Parlora.BackendRequest.md): A uniquely identified revision of application-owned work. - [Parlora.BackendResult](Parlora.BackendResult.md): Verified backend outcome; cancellation does not imply reversal of side effects. - [Parlora.Channel.Capabilities](Parlora.Channel.Capabilities.md): Effective channel formats, controls, and sources of playback evidence. - [Parlora.ContextDecision](Parlora.ContextDecision.md): An application proposal validated against the current context revision. - [Parlora.ContextSnapshot](Parlora.ContextSnapshot.md): Immutable context supplied to the application policy, not a persistence format. - [Parlora.ContextTrigger](Parlora.ContextTrigger.md): The observation that requests context interpretation. - [Parlora.ContextUpdate](Parlora.ContextUpdate.md): Concise provider context with command and task correlation. - [Parlora.Limits](Parlora.Limits.md): Finite per-session bounds. Audio capacities are bytes of the agreed pipeline format. - [Parlora.Provider.Capabilities](Parlora.Provider.Capabilities.md): Effective voice-provider features; advertise only implemented behavior. - Session API - [Parlora](Parlora.md): An Elixir framework for building real-time voice AI agents. - [Parlora.Event](Parlora.Event.md): A live observation, independent of business completion or confirmed playback. - [Parlora.SessionHealth](Parlora.SessionHealth.md): Content-free observations in `Parlora.status/1` under `:health`. - [Parlora.Status](Parlora.Status.md): Public map returned by `Parlora.status/1`; this module defines types only. - Channels - [Parlora.Channels.Browser](Parlora.Channels.Browser.md): One browser audio connection, terminated at Parlora with ExWebRTC and libopus. - [Parlora.Channels.Browser.Signaling](Parlora.Channels.Browser.Signaling.md): Mountable WebSock signaling for one application-owned session. - [Parlora.Channels.LocalAudio](Parlora.Channels.LocalAudio.md): CPAL microphone and speaker audio, converted through Rubato to mono PCM16/24 kHz. - [Parlora.Channels.Telnyx](Parlora.Channels.Telnyx.md): Incoming Telnyx bidirectional Media Streaming with raw mono PCMU at 8 kHz. - [Parlora.Channels.Telnyx.Commands](Parlora.Channels.Telnyx.Commands.md): Bounded incoming-call commands over Mint. The host supplies `:api_key`. - [Parlora.Channels.Telnyx.Endpoint](Parlora.Channels.Telnyx.Endpoint.md): Mountable signed Telnyx v2 webhook admission. This module starts no listener. - [Parlora.Channels.Telnyx.Signature](Parlora.Channels.Telnyx.Signature.md): Ed25519 verification of Telnyx v2 webhooks using the original body bytes. - [Parlora.Channels.Twilio](Parlora.Channels.Twilio.md): Bidirectional Twilio Media Streams with raw mono μ-law at 8 kHz. - [Parlora.Channels.Twilio.Endpoint](Parlora.Channels.Twilio.Endpoint.md): Mountable, signed incoming-call and WebSocket routes. No listener is started. - [Parlora.Channels.Twilio.Signature](Parlora.Channels.Twilio.Signature.md): Twilio HMAC-SHA1 request verification using an explicit public URL. - [Parlora.Channels.Twilio.Socket](Parlora.Channels.Twilio.Socket.md): WebSock transport for an authenticated and authorized Twilio call. - Providers - [Parlora.Providers.Gemini](Parlora.Providers.Gemini.md): Gemini 3.8 Live with application-owned credentials and backend work. - [Parlora.Providers.Grok](Parlora.Providers.Grok.md): Grok Voice 3.0 with application-owned backend work over WebSocket. - [Parlora.Providers.OpenAI](Parlora.Providers.OpenAI.md): GPT-Live-1 client delegation over Mint.WebSocket, with explicit application-owned credentials. - Backends - [Parlora.Backends.Jido](Parlora.Backends.Jido.md): Runs a fresh Jido AgentServer for each request, observed outside session/audio callbacks. - [Parlora.Backends.Legion](Parlora.Backends.Legion.md): Optional Legion 0.6.0 execution in isolated Lua agents. - [Parlora.Backends.Sagents](Parlora.Backends.Sagents.md): Runs one isolated, bounded Sagents execution per admitted request. - Adapter contracts - [Parlora.Backend](Parlora.Backend.md): Contract for application-owned work, independent of the voice provider. - [Parlora.Channel](Parlora.Channel.md): Participant audio and playback controls, independent of device or transport. - [Parlora.ContextPolicy](Parlora.ContextPolicy.md): Interprets conversation and application state into backend task decisions. - [Parlora.Provider](Parlora.Provider.md): Continuous full-duplex voice, client delegation and correlated context updates. - Exceptions - [Parlora.Error](Parlora.Error.md): A sanitized failure at a Parlora boundary.