跳转至

apogee-ai-voice

STT, TTS, VAD, barge-in and hot-word detection for voice agents.

This page is being expanded

The tables below are generated from the package source, so they are accurate. Narrative guides and worked examples are being written module by module.

Public API

Everything below is exported from the package root, so from apogee_ai_voice import X works for each name.

Layer Symbols
Application · DTOs GenerateImageDTO, GenerateVideoDTO, ListVoicesDTO, PipelineTurnDTO, SynthesizeDTO, TranscribeDTO
Application · Use cases GenerateImageUseCase, GenerateVideoUseCase, ListVoicesUseCase, RunVoicePipelineUseCase, SynthesizeUseCase, TranscribeUseCase
Domain AudioChunk, AudioFormat, AudioFormatSpec, ImageAsset, ImageGenUnavailable, ImageRequest, ImageResult, ImageStyle, LanguageCode, LatencyBudget, Modality, RealtimeEvent, RealtimeSession, SynthesisRequest, SynthesisResult, Transcript, TranscriptSegment, VadDecision, VadEvent, VideoGenUnavailable, VideoModel, VideoRequest, VideoResult, Voice, VoiceProfile
Domain · Enums RealtimeEventKind
Domain · Exceptions LanguageNotSupportedException, RealtimeError, STTUnavailableException, TTSUnavailableException, TransportError, VoiceError
Domain · Protocols (ports) IImageGen, IRealtimeVoice, ISTT, ITTS, ITransport, IVideoGen, IVoiceActivityDetector
Infrastructure AssemblyAISTT, CartesiaTTS, DeepgramSTT, DeepgramTTS, EchoImageGen, EchoSTT, EchoTTS, EchoVideoGen, ElevenLabsTTS, EnergyVAD, FluxGen, ImagenGen, LiveKitTransport, LocalAudioTransport, OpenAIImageGen, OpenAIRealtimeAdapter, OpenAITTS, PipecatRunStats, PipecatStyleRuntime, RunwayVideoGen, SoraVideoGen, StableDiffusionGen, VeoVideoGen, VoiceRegistry, WhisperSTT

Install

Bash
pip install apogee-ai-voice

The base install has no required third-party dependency. Each extra unlocks one adapter.

Extra Pulls in
elevenlabs elevenlabs>=1.5
cartesia cartesia>=1.0
openai openai>=1.30
deepgram deepgram-sdk>=3.7
assemblyai assemblyai>=0.30
livekit livekit>=0.18, livekit-agents>=0.10
runway runwayml>=2.0
all elevenlabs>=1.5, cartesia>=1.0, openai>=1.30, deepgram-sdk>=3.7, assemblyai>=0.30, livekit>=0.18, livekit-agents>=0.10, runwayml>=2.0

CLI

Installed as a console script, and registered as an apogee plugin so the same subcommands work under the global binary.

Bash
apogee-voice --help