We are releasing Duplex Cue today - a new evaluation dataset of in-turn adaptation behavior in full-duplex voice agents. Full-duplex evaluation often emphasizes whether an agent keeps speaking or stops during speech overlaps. That binary cannot express a third response humans use routinely: continuing to speak while incorporating what the listener just contributed. A full-duplex agent can keep talking through the cue, incorporate it without stopping, or hand over the turn. A simple stop-or-continue score cannot tell those behaviors apart. We introduce Duplex Cue, an evaluation of this in-turn adaptation behavior in full-duplex voice agents. Duplex Cue separates listener intent (backchannel, collaboration, or interruption) from speaker behavior: continuing unchanged, adapting within the turn, or yielding. Adaptation includes acknowledgment as well as content revision. In a single-model case study using 300 human-confirmed cues from unscripted English conversations, we compare recorded human responses with @nvidia 's PersonaPlex continuations generated while replaying the listener's audio. We retain 208 pairs with the ongoing speaker active at cue onset and a scorable response in each condition. For backchannels, PersonaPlex and recorded speakers show similar response patterns. For interruptions, PersonaPlex yields more often and adapts less often than recorded speakers. On the 66 collaboration pairs, recorded speakers adapt in 68.2% of cases, compared with 34.8% for PersonaPlex. The model otherwise continues unchanged (42.4%) or yields (22.7%). Natural human interaction is much more complicated than a simple stop-or-continue, and we are only scratching the surface of it! Findings, paper, and public dataset can be found at

We published how @thinkymachines 's Inkling did on Voice Code Bench few weeks ago, and then we asked, what if we fine-tune Inkling? Inkling's main value prop is its small size and robustness, making it a good base model for domain specific fine-tuning. So we that's what we did. We fine-tuned the same Inkling speech model on 1, 25, and 100 hours of our proprietary data containing alphanumeric entities. And the curve kept moving. 🍊 Not surprising, on the standard 300-item VoiceCodeBench evaluation, the 100-hour checkpoint delivered the strongest result. Untrained Inkling → 100-hour post-trained Inkling: 📈 Task Success Rate: 56.33% → 79.00% (+22.67 pts) 📈 Entity recovery (CTEM): 86.84% → 94.80% (+7.96 pts) 📉 VoiceCodeBench WER: 2.3748% → 1.6107% (32.2% relative reduction) 🔧 139 misses fixed, 21 prior hits regressed: net +118 exact entities recovered And the gains scaled as data scaled: 1h: 88.06% CTEM / 59.33% TSR / 2.8188% VCB WER 25h: 92.85% / 72.67% / 1.8517% 100h: 94.80% / 79.00% / 1.6107% The 100-hour model recovered values the base model missed: •--revert-last, instead of splitting one flag into “--revert --last” • tests/auth/login.spec.ts, instead of test/auth/login.spec.ts • ALLOWLIST_CIDR, instead of inserting an extra underscore • SN-7KX-9042, instead of dropping the final digit The largest entity-type gains were email addresses (+30.8 pts), postal addresses (+30.0), file paths (+23.5), environment variables (+22.9), and IP addresses (+20.0). That’s the Besimple thesis: targeted human data can move the production metrics that matter for voice agents, even when the base model is already strong. DM me if you want to evaluate your model on this benchmark or build the data that moves it. 🍊 #SpeechRecognition# #VoiceAI# #ASR# #Transcription# #DataQuality# #PostTraining# #Benchmarks# Checkout the full blog at: