The whole voice stack, on our own models.

Five stages, our own models.

Audio comes in, speech goes out. Psst, Turno and Versa are ours; recognition runs NVIDIA’s open model on our servers until our own replaces it; the language model is a provider’s.

• PsstSpectral-gating noise suppression, run as middleware ahead of transcription. Isolates speech in cafes, cars, wind and open offices. Ours.
• TurnoTurn-taking and barge-in decided on the raw audio stream — a real interruption versus a “mm-hmm” — instead of silence-timeout heuristics. 18,651 parameters. Ours.
• ConvSpeech-to-text. Conv 1.0 runs NVIDIA’s open Parakeet TDT 0.6B on our own servers, with no audio sent to a third-party cloud. Conv 2.0, our own streaming recogniser, is in development. NVIDIA’s model today, ours next.
• LLMBest-in-class low-latency providers, pluggable, orchestrated by our pipeline rather than stitched together client-side. Provider.
• VersaNon-autoregressive speech synthesis built on flow matching — parallel generation instead of token-by-token. Nine languages, 10 voices, zero-shot voice cloning, 91.4 million deployable parameters. First audio in a median of 64 ms; on CPU-only servers, until September 2026, 139 ms. Ours.

More on our models →

Everything around the models.

Agents, phone numbers, a speech studio, transcription, analytics. All on app.lokutor.com, all on the same minutes.

app.lokutor.com

Agents

Build an agent in the dashboard.

Prompt, voice, language, tools and cross-call memory in one form. Talk to it in the browser before it ever takes a call.

• 10 voices, nine languages
• Tool calling and memory
• Test it live, in the browser

app.lokutor.com

Phone

Give it a number.

Provision phone numbers, take inbound calls, place outbound ones, and run bulk campaigns from the same agent.

• Inbound and outbound
• Single and bulk dialing
• Numbers $5 a month

app.lokutor.com

Speech

A speech studio, not just an endpoint.

Type, pick a voice, hear it. Streaming synthesis with visemes for lip-sync, and voice cloning.

• Streams from the first chunk
• Visemes included
• $4.10 per million characters

app.lokutor.com

Transcription & noise

Conv and Psst on their own.

Transcribe audio and clean it up as standalone endpoints, the same models the agent uses, billed by the second from the same balance.

• Speech-to-text in six languages
• Noise suppression as middleware
• One key, one bill

app.lokutor.com

Analytics & logs

See every conversation.

Transcripts, per-turn latency, minutes used and errors, per agent and per call, so you know what the agent said and how fast.

• Transcripts per call
• Latency per turn
• Usage and cost per call

A voice agent in ten lines.

JavaScript and Python SDKs, a plain REST endpoint for speech, and a Pipecat service if you already have a pipeline.

import { VoiceAgentClient } from '@lokutor/sdk';

const agent = new VoiceAgentClient({
  apiKey: 'YOUR_API_KEY',
  prompt: 'You are the support agent for Acme.',
  voice: 'F1',
  language: 'en',
  serverUrl: 'wss://api.lokutor.com/ws/agent',
});

await agent.startManaged();

import requests

r = requests.post(
    "https://api.lokutor.com/tts/synthesize",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={"text": "Hello from Lokutor.", "voice": "F1", "language": "en"},
)
open("audio.wav", "wb").write(r.content)

curl -X POST "https://api.lokutor.com/tts/synthesize" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello from Lokutor.", "voice": "F1", "language": "en"}' \
  --output audio.wav

# pip install pipecat-lokutor
from pipecat_lokutor import LokutorTTSService

tts = LokutorTTSService(api_key="YOUR_API_KEY", sample_rate=44100)
# use it as the TTS stage of any Pipecat pipeline

Full documentation →