Voice · Streaming
The real-time audio pipe for production voice AI.
Fork live call audio over a bidirectional WebSocket to your speech-to-text, language model and text-to-speech with sub-second round trip. Works from PSTN, SIP and WebRTC calls, in the cloud or on-premise, with any agent framework you already use, and with the EnableX AI Voice Agent when you want managed.
Stream your first call to your AI
On a connected call, send one request with your voice_id and WSS host, and EnableX opens a bidirectional audio pipe to your STT, LLM, and TTS. Copy, paste, run.
curl -X PUT https://api.enablex.io/voice/v1/call/$voice_id/stream \
-u APP_ID:APP_KEY \
-H "Content-Type: application/json" \
-d '{ "wss_host": "wss://your-host:port" }'const axios = require('axios');
const voiceId = 'YOUR_VOICE_ID';
const res = await axios.put(
`https://api.enablex.io/voice/v1/call/${voiceId}/stream`,
{ wss_host: 'wss://your-host:port' },
{ auth: { username: 'APP_ID', password: 'APP_KEY' } }
);
console.log(res.data);import requests
from requests.auth import HTTPBasicAuth
voice_id = "YOUR_VOICE_ID"
res = requests.put(
f"https://api.enablex.io/voice/v1/call/{voice_id}/stream",
auth=HTTPBasicAuth("APP_ID", "APP_KEY"),
json={"wss_host": "wss://your-host:port"}
)
print(res.json())Everything a production voice AI agent needs
Stream audio and plug your STT, LLM, TTS, or hand off to the managed EnableX AI Voice Agent. Both paths first-class.
Bidirectional WebSocket
Receive caller audio and send synthesised audio back, both directions, base64-encoded mu-law 8 kHz mono, standard telephony audio.
Sub-second round-trip
Tier-1 routes deliver end-to-end round-trip sub-second: the difference between an agent that feels human and one that feels like a bot.
Bring your own stack
Works with open-source agent frameworks and any STT, LLM or TTS provider. Plug your stack, we move the audio.
Speech recognition built in
30+ languages including 13 Indian languages, with code-switching, or bring your own engine.
PSTN, SIP, WebRTC sources
Stream from inbound PSTN calls, SIP trunks, or WebRTC sessions. Same WebSocket endpoint, same payload format.
Cloud, hybrid, on-prem
Run inside your datacentre for sovereign AI and voice biometrics, for regulated voice workloads that must stay in-country.
What teams build with Voice Streaming
From a phone number to a production voice AI agent on one stack.
AI voice agents
STT → LLM → TTS pipelines that handle inbound and outbound calls end-to-end. Pipecat, LiveKit, or your own orchestrator.
AI voice agentsVoice biometrics & auth
Verify caller identity in the first seconds of a call. On-prem deployment keeps voiceprints inside your data centre.
Voice AI for bankingReal-time call analytics
Live transcription, sentiment, intent, and compliance scoring as the call happens, not hours later.
Voice AI for insuranceTele-triage with AI
Stream patient calls to clinical NLP for symptom triage, appointment routing, and structured EHR capture. HIPAA-ready controls.
Voice AI for healthcareSupervisor whisper & coaching
Live listen-in, real-time keyword alerts, AI-suggested next-best-utterance, designed to surface risk in the moment.
Talk to salesReal-time recording & archive
Stream both legs to your archive in real time, with encryption in transit and at rest and retention rules per region.
Talk to salesSub-second. Bidirectional. Production-grade.
We built the pipe for what voice AI needs: bidirectional, low-jitter, reconnect-safe, with hooks into the EnableX AI Voice Agent if you want managed instead of DIY.
- Base64 mu-law 8 kHz mono: the same telephony format your ASR engine already speaks.
- Bidirectional: receive caller audio, send synthesised audio back. Open-source agent frameworks work out of the box.
- Sub-second round-trip on tier-1 routes: the threshold where conversation feels natural.
- EnableX handles WebSocket reconnect, sequencing, clock-sync, and DTMF eventing; you focus on the agent.
- On-prem deployment for sovereign AI, voice biometrics, and regulated healthcare workloads.
Why teams build voice AI on EnableX
What sets the EnableX Voice Streaming API apart.
On-prem & sovereign-AI ready
Our cloud, your cloud, private cloud or fully on-premise, including the speech models, for voice biometrics, sovereign AI mandates and regulated workloads.
Speech recognition built in
30+ languages including 13 Indian languages, with code-switching, or bring your own engine on the same pipe.
DIY pipe + managed AI Agent
Use the WebSocket directly with your stack, or hand off to the managed EnableX AI Voice Agent: same platform and billing. Both paths supported.
Where teams use the Voice Streaming API
Pay-as-you-stream. No platform fee.
Voice Streaming API FAQs
Build with Voice Streaming API: Developer Docs
See EnableX in action.
Talk to sales or start a free trial. No credit card required.
Free trial credits · No credit card · API keys in 2 minutes