Performance Benchmarks

Real production measurements, not marketing rounding: where the time goes in a turn, and how each of our models compares, with the methodology to reproduce it.

The Full Pipeline Latency

In real-world voice AI, TTS latency alone doesn't tell the story. We measure time to first audio: from the moment the user stops speaking to the first sound of the reply. The agent streams token by token, so the caller never waits for a whole sentence — only for the first short chunk.

Where the time goes

Measured on production on 10 October 2026: 30 calls, one at a time, from a machine in the same AWS region as our servers (Ireland), so your own internet connection is not in the figure. Each call says one Spanish sentence after the agent's greeting; the stages come from our servers' own log of each of those turns. Hover a segment.

The language model starts on the caller's words before they have finished; on 26 of the 30 calls its reply was ready by the end of the wait, and on the other 4 the caller waited 925–1,137 ms. Published guidance treats under 800 ms as conversationally natural; an independent survey places production voice-AI platforms at a 1,400–1,700 ms median. Streaming text to speech on its own reaches the first audio in 65 ms (median of 40 requests on an open connection; about 200 ms for the first request on a new one). In September 2026, when everything but the language model still ran on CPU-only servers (AWS c8g.2xlarge, Graviton4), callers waited 385–442 ms in the best case and about 660 ms typically, and Versa's first audio took a median of 139 ms: on CPUs alone, already in real time.

Prices

About 5.2¢ a minute for a voice agent at a typical prompt: 4.2¢ for the platform, with speech recognition, the voice, turn-taking and orchestration inside it, plus the language model by the token. Prepaid and billed by the second; there are no plans.

Phone numbers $5 a month each; every line and country on the pricing page. Dedicated and private deployments are quoted separately.

Noise Suppression — Psst

Psst is our proprietary strictly-causal noise suppression model, running as low-latency middleware ahead of speech recognition. We score it on downstream word-error rate — the only metric that reaches the caller — not on perceptual audio quality, which turned out to point the wrong way entirely.

300 LibriSpeech test-clean utterances mixed with held-out noise and interfering speakers; no evaluation audio appears anywhere in training. Psst leads on downstream word error by 36% over DeepFilterNet3, the strongest published real-time CPU denoiser, at comparable cost and inside a 32ms strictly-causal budget DeepFilterNet3 doesn't meet. Against competing speech — other people talking nearby, the most common real-world complaint — Psst recovers 14–28% of word errors. It runs 18 concurrent call legs per CPU core at 2.7MB, so noise suppression adds no meaningful capacity cost.

The word-error and perceptual rankings invert completely. DeepFilterNet3 leads PESQ by 0.70 and SI-SDR by 8dB and is the worst system here for transcription — worse than no denoising at all. Psst is the only system tested that improves on doing nothing. Optimizing for the industry-standard perceptual metric would have degraded the product.

Turn-Taking — Turno

Turno, our turn-taking and barge-in model, is 18,651 parameters. On a neutral real-world test set it beats Silero VAD on Average Precision (0.931 vs 0.876) and holds that precision flat from clean audio down to 0dB signal-to-noise, at roughly half Silero's per-inference cost (0.083ms per frame). TEN-VAD is ahead on voice activity alone (0.985); Turno also predicts the end of a turn and tells a real interruption from a backchannel, which voice-activity models do not. Retraining on real echo data cut the real-device echo false-confirm rate from 93.0% to 11.4%.

Speech Recognition — Conv

Conv 1.0 serves production today: NVIDIA’s open Parakeet TDT 0.6B model, served on our own GPUs with no cloud fallback — about 45 ms to transcribe a four-second sentence, and around 36 utterances per second per card under sustained load.

Conv 2.0, our own streaming recogniser and our next release: 450 million parameters, a quarter smaller than Parakeet 0.6B. Scored with the Artificial Analysis streaming benchmark protocol (AA-WER Streaming) and its published normalisation, on our own harness; the conversational set is our public stand-in for Artificial Analysis's held-out agent and customer set.

Why Small Models Matter

State-of-the-art voice AI today means giant models on giant GPU fleets. That model breaks down the moment voice AI needs to live outside a data center.

• Lifetime cost: sell a voice-enabled device once, and cloud-only inference means paying for it for that device's entire lifetime. A model small enough to run on the device removes that tax.
• Latency & connectivity: round-trips to a GPU cloud fight against real-time conversation. Running on the processor already in the device, or on your own server, removes a hop.
• Privacy & compliance: for regulated buyers, audio that never has to leave the customer's environment shortens procurement under GDPR and the EU AI Act — compliance as a distribution advantage, not a cost center.

Verify Data

Want to run these benchmarks on your own data? Contact us for a technical consultation or access to our evaluation scripts.