jargo

package module

v0.1.0 Latest Latest Go to latest Published: Aug 30, 2026 License: BSD-2-Clause Imports: 0 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/gojargo/jargo

Links

Open Source Insights

README ¶

A WebRTC-native, audio-first conversational-AI framework for Go.

Go version

jargo is a framework for real-time voice agents in Go: audio in over WebRTC, a streaming transcription → reasoning → speech pipeline with turn-taking and barge-in, and audio back out.

[!WARNING] Early work in progress. Not ready for production. The public API is unstable and changes in any release. Pin an exact version and read CHANGELOG.md before upgrading.

Why?

Pipecat is great, and jargo is a port of it. The architecture and many design decisions are Pipecat's.

Python might not be the way

This port exists for one reason: I'd rather not run a voice agent on Python.

Python is the right tool when you need the AI/data-science ecosystem. A real-time voice server doesn't: the models run as services or as ONNX, and what's left is plumbing: audio framing, WebRTC, concurrency, and shipping a binary. For that, Go is a better fit: one static binary to deploy, low and predictable memory, fast startup, and real concurrency for many simultaneous sessions without a GIL. The heavy numerics stay where they belong (the ONNX Runtime, the remote services), so giving up Python costs little here. See the benchmarks for the honest performance picture.

Features

WebRTC, pure Go (Pion): audio in and out of the browser.
Opus, pure Go encode + decode via pion/opus; C libopus optional with -tags libopus.
Resampling, pure Go via go-resample; libsoxr optional with -tags libsoxr.
Streaming voice pipeline: STT → LLM → TTS, with prompt caching.
Speech-to-speech: single-model voice agents (OpenAI Realtime, Gemini Live, AWS Nova Sonic).
Turn-taking & barge-in: Silero VAD + Smart Turn v3, local ONNX.
Telephony (optional): inbound/outbound phone calls over Twilio Media Streams.
User-idle watchdog: re-engage or hang up when the caller goes silent.
RTVI data channel: works with existing RTVI clients.
Pluggable services: swap any STT/LLM/TTS behind a small interface.
Concurrent by design: independent processors; interruptions are frames.

Providers

Pick any per category; each is a small Config + constructor.

STT: Deepgram, AssemblyAI, Gladia, Speechmatics, Soniox, Whisper (OpenAI/Groq/local), Azure, xAI, ElevenLabs, Cartesia, NVIDIA.
LLM: Anthropic (direct + Bedrock), OpenAI (chat + Responses), Google Gemini (direct + Vertex), Groq, Together, Fireworks, DeepSeek, Cerebras, Perplexity, OpenRouter, xAI, Ollama, NVIDIA, Mistral, Nebius, SambaNova, Qwen, Azure OpenAI.
TTS: ElevenLabs, Cartesia, Rime, LMNT, Kokoro, Piper, Pocket TTS, Deepgram, OpenAI, Azure, Hume, Fish, MiniMax, xAI, NVIDIA, Soniox.
Speech-to-speech: OpenAI Realtime (direct + Azure), Gemini Live (direct + Vertex), AWS Nova Sonic, xAI Realtime.
Memory: mem0.

Usage

go get github.com/gojargo/jargo

A bot is an STT → LLM → TTS pipeline over a WebRTC transport. The heart of it:

stt := chat.NewSTT(chat.STTConfig{APIKey: key, SampleRate: opus.SampleRate})
llm := chat.NewLLM(chat.LLMConfig{APIKey: key})
tts := chat.NewTTS(chat.TTSConfig{APIKey: key})

t := rtc.NewTransport(conn, transport.DefaultParams())
agg := aggregators.New(frames.NewLLMContext("You are a helpful voice assistant."))

task := pipeline.NewWorker(pipeline.New(
	t.Input(), stt, agg.User(), llm, tts, t.Output(), agg.Assistant(),
), pipeline.WorkerConfig{})
task.Run(ctx)

examples/voice/openai is that pipeline as a complete server (WebRTC signaling, VAD/turn-taking, barge-in).

Run it in Docker: build on the gojargo/jargo-build base and ship on the distroless gojargo/jargo runtime (it bundles the ONNX Runtime), then:

docker run --rm -p 8080:8080 -e OPENAI_API_KEY=$OPENAI_API_KEY my-bot

See Deploy with Docker for the Dockerfile and the Quickstart for the full setup.

Examples

Runnable bots live in examples/:

echo: hear yourself back, no API keys.
voicebot: the full voice agent (STT → LLM → TTS over WebRTC) with turn-taking, long-term memory, and tracing.
voice/: one headless backend per provider, each wiring its STT/LLM/TTS explicitly and exposing the WebRTC /offer endpoint (no web UI). Run with go run ./examples/voice/<provider> (e.g. deepgram, cartesia, openai) and drive it from a browser client, the nextjs-voicebot in jargo-client-react.
twiliobot: a phone agent over Twilio Media Streams, with the idle watchdog.

The fastest way to try them (locally or with Docker) is the Quickstart.

go run ./examples/echo                 # then open http://localhost:8080

Documentation

gojargo.github.io/jargo is the full documentation. The same pages live in docs/ and read fine on GitHub.

Start with Architecture for the model, or Frames and Processors for the engine. Writing a processor covers extending it. The API reference is the Go reference.

Dependencies

The default build is cgo-free: CGO_ENABLED=0 go build ./... works with no C toolchain. Two native runtimes are still used, but bound through purego and loaded at run time, so they need their shared library present at runtime and nothing at build time:

ONNX Runtime: VAD + end-of-turn detection (JARGO_ONNXRUNTIME_LIB).
RNNoise: optional input noise reduction (JARGO_RNNOISE_LIB).

Opus and resampling are pure Go by default; the C libopus (-tags libopus) and libsoxr (-tags libsoxr) are the only cgo in the tree, and both are optional. The base images bundle all of them.

License & attribution

jargo is a Go port of Pipecat, distributed under the same BSD 2-Clause License. The upstream copyright (Copyright (c) 2024–2026, Daily) is preserved verbatim in LICENSE; see NOTICE for details. jargo is an independent project, not affiliated with or endorsed by Daily.

Documentation ¶

Overview ¶

Package jargo is a framework for building real-time voice and multimodal AI pipelines.

Source Files ¶

View all Source files

doc.go

Directories ¶

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL

Path	Synopsis
adapter Package adapter converts jargo's universal conversation context into the request each LLM provider takes.	Package adapter converts jargo's universal conversation context into the request each LLM provider takes.
anthropic Package anthropic converts a universal conversation into the request Anthropic takes, and with it Bedrock, which serves the same models.	Package anthropic converts a universal conversation into the request Anthropic takes, and with it Bedrock, which serves the same models.
gemini Package gemini converts a universal conversation into the request Gemini takes, and with it Vertex AI, which serves the same models.	Package gemini converts a universal conversation into the request Gemini takes, and with it Vertex AI, which serves the same models.
mistral Package mistral converts a universal conversation into the request Mistral takes.	Package mistral converts a universal conversation into the request Mistral takes.
openai Package openai converts a universal conversation into the chat-completions request OpenAI takes, and with it every endpoint that speaks OpenAI's API.	Package openai converts a universal conversation into the chat-completions request OpenAI takes, and with it every endpoint that speaks OpenAI's API.
perplexity Package perplexity converts a universal conversation into the request Perplexity takes.	Package perplexity converts a universal conversation into the request Perplexity takes.
realtime Package realtime converts a universal conversation into what OpenAI's Realtime API takes on a session.	Package realtime converts a universal conversation into what OpenAI's Realtime API takes on a session.
responses Package responses converts a universal conversation into the request OpenAI's Responses API takes, over HTTP and over the WebSocket alike: the two transports carry the same request.	Package responses converts a universal conversation into the request OpenAI's Responses API takes, over HTTP and over the WebSocket alike: the two transports carry the same request.
audio Package audio defines the interfaces for pluggable audio processing a transport applies to the media it sends and receives.	Package audio defines the interfaces for pluggable audio processing a transport applies to the media it sends and receives.
dtmf Package dtmf generates the dual-tone multi-frequency signals a telephone keypad produces, as 16-bit mono PCM.	Package dtmf generates the dual-tone multi-frequency signals a telephone keypad produces, as 16-bit mono PCM.
g711 Package g711 implements the ITU-T G.711 companding codecs used by telephony media streams (Twilio, Telnyx, Plivo and the wider PSTN): μ-law (PCMU), the North American variant, and A-law (PCMA), used elsewhere.	Package g711 implements the ITU-T G.711 companding codecs used by telephony media streams (Twilio, Telnyx, Plivo and the wider PSTN): μ-law (PCMU), the North American variant, and A-law (PCMA), used elsewhere.
loudness Package loudness measures how loud audio is, to ITU-R BS.1770 (the EBU R128 standard), and normalizes the result to a 0..1 scale.	Package loudness measures how loud audio is, to ITU-R BS.1770 (the EBU R128 standard), and normalizes the result to a 0..1 scale.
mixer Package mixer mixes background audio into a transport's outgoing audio.	Package mixer mixes background audio into a transport's outgoing audio.
noise/gate Package gate provides a noise gate: an audio.Filter that silences a chunk of audio whose RMS energy falls below a threshold, suppressing low-level background hum and hiss between speech.	Package gate provides a noise gate: an audio.Filter that silences a chunk of audio whose RMS energy falls below a threshold, suppressing low-level background hum and hiss between speech.
noise/rnnoise Package rnnoise is an input audio filter that suppresses background noise with RNNoise, Xiph's recurrent-network denoiser.	Package rnnoise is an input audio filter that suppresses background noise with RNNoise, Xiph's recurrent-network denoiser.
onset Package onset detects the first audible sample in a stream of PCM audio.	Package onset detects the first audible sample in a stream of PCM audio.
opus Package opus wraps the Opus codec for jargo's audio path: it decodes Opus packets to PCM on the way in and encodes PCM to Opus packets on the way out.	Package opus wraps the Opus codec for jargo's audio path: it decodes Opus packets to PCM on the way in and encodes PCM to Opus packets on the way out.
resample Package resample converts interleaved S16LE PCM audio between sample rates, preserving the channel count.	Package resample converts interleaved S16LE PCM audio between sample rates, preserving the channel count.
turn Package turn provides end-of-turn detection: it decides when a user has actually finished speaking, as opposed to merely pausing.	Package turn provides end-of-turn detection: it decides when a user has actually finished speaking, as opposed to merely pausing.
vad Package vad provides voice activity detection: it tells the pipeline when a user is speaking.	Package vad provides voice activity detection: it tells the pipeline when a user is speaking.
vad/controller Package controller drives a voice-activity detector and reports what it hears.	Package controller drives a voice-activity detector and reports what it hears.
bus Package bus is the pub/sub messaging that connects workers to each other and to the runner that manages them.	Package bus is the pub/sub messaging that connects workers to each other and to the runner that manages them.
clock Package clock provides the timing source a pipeline uses for presentation timestamps and elapsed-time measurements.	Package clock provides the timing source a pipeline uses for presentation timestamps and elapsed-time measurements.
cmd
jargo command Command jargo is the command-line entry point for jargo tooling.	Command jargo is the command-line entry point for jargo tooling.
eval Package eval is a behavioral eval harness for jargo bots.	Package eval is a behavioral eval harness for jargo bots.
examples
echo command Command echo is a WebRTC echo bot built on jargo: it receives microphone audio from a browser, runs it through a pipeline, and sends it back so you hear yourself.	Command echo is a WebRTC echo bot built on jargo: it receives microphone audio from a browser, runs it through a pipeline, and sends it back so you hear yourself.
eval command Command eval is a tiny demo bot for the jargo eval harness.	Command eval is a tiny demo bot for the jargo eval harness.
flows command Command flows is a headless jargo voice backend driven by a conversation flow.	Command flows is a headless jargo voice backend driven by a conversation flow.
localaudio command Command localaudio echoes the local microphone straight back to the local speaker using jargo's pure-Go local-audio transport.	Command localaudio echoes the local microphone straight back to the local speaker using jargo's pure-Go local-audio transport.
silktest command Throwaway: exercise jargo's audio/opus Encoder end-to-end, the way the transport does — feeding 20 ms frames of 48 kHz mono S16LE PCM and decoding the packets back.	Throwaway: exercise jargo's audio/opus Encoder end-to-end, the way the transport does — feeding 20 ms frames of 48 kHz mono S16LE PCM and decoding the packets back.
twilio/outbound command Command outbound is an outbound telephony voice agent: it places a phone call over Twilio, talks to whoever answers, collects a few details and hangs up.	Command outbound is an outbound telephony voice agent: it places a phone call over Twilio, talks to whoever answers, collects a few details and hangs up.
twiliobot command Command twiliobot is a telephony voice agent built on jargo: a Twilio phone call streams audio over a WebSocket (μ-law 8 kHz), an STT service transcribes it, an LLM reasons over it, a TTS service speaks the reply, and the audio goes back to the caller.	Command twiliobot is a telephony voice agent built on jargo: a Twilio phone call streams audio over a WebSocket (μ-law 8 kHz), an STT service transcribes it, an LLM reasons over it, a TTS service speaks the reply, and the audio goes back to the caller.
voice/anthropic command Command voice-anthropic is a headless jargo voice backend.	Command voice-anthropic is a headless jargo voice backend.
voice/assemblyai command Command voice-assemblyai is a headless jargo voice backend.	Command voice-assemblyai is a headless jargo voice backend.
voice/azure command Command voice-azure is a headless jargo voice backend.	Command voice-azure is a headless jargo voice backend.
voice/cartesia command Command voice-cartesia is a headless jargo voice backend.	Command voice-cartesia is a headless jargo voice backend.
voice/deepgram command Command voice-deepgram is a headless jargo voice backend.	Command voice-deepgram is a headless jargo voice backend.
voice/deepseek command Command voice-deepseek is a headless jargo voice backend.	Command voice-deepseek is a headless jargo voice backend.