Skip to content

/

ZeroHour

ZeroHour aggregates public sources — security and AI press, vendor advisories, NVD, CISA KEV, FIRST EPSS, leak-site trackers and Have I Been Pwned — and classifies them with AI. Summaries can be wrong: read the original. All times are UTC.

News CVEs Incidents Tags Daily brief Traffic Sources RSS & API Kiosk Health Sitemap

Home
News feed
Exploited now

Security

CVEs
Ransomware & breaches
Threat actors & entities
IOCs

AI

AI & models

Explore

Tags
Daily brief
Analytics
Traffic & readers
Ask the feed
Sources

You

Watchlist
Saved
RSS, API & alerts

07:57 UTC· 9m ago

#speech — news, CVEs and analysis · ZeroHour

today 46 visits · 59 views · 7 reads·all time 17K views

Tag

#speech

6 stories this week · 11 this month · 12 in 6 months · mostly AI

stories per day · 30 days

Related#full-duplex 4 #benchmark 3 #voice-agents 3 #conversational-ai 2 #evaluation 2 #gpt-live-1 2 #multimodal 2 #openai 2 #open-weights 2 #turn-taking 2 #alibaba 1 #api 1 #asr 1 #audio-language-models 1 #chinese 1 #contrastive-learning 1 #dialogue 1 #diarization 1

12 stories, newest first · one card per story

arXiv cs.AI / cs.LG / cs.CL·2d ago

To Trust or Not to Trust: Retrieval-Augmented Fact Checking in Speech

#verispeak #fact-checking #speech

AI research

arXiv cs.AI / cs.LG / cs.CL·2d ago

Do Audio Language Models Hear and Read Distinctive Features Alike?

#audio-language-models #qwen2.5-omni #phonology

AI research

MarkTechPost·3d ago

NVIDIA Releases Nemotron 3 Diarization: A 100M-Parameter Open-Weight Model That Tracks 8 Speakers in Real Time

#nvidia #nemotron #diarization

2 sources 5 min

The Decoder·3d ago

Alibaba launches Qwen Audio 3.1 with new models and slashes AI audio prices by up to 95 percent

#alibaba #qwen #qwen-audio

MarkTechPost·4d ago

Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning

#kyutai #voice-of-reason #glm-4-voice

4 min

arXiv cs.AI / cs.LG / cs.CL·5d ago

ToneCL: Contrastive Learning for Few-Shot Syllable-Level Tone Classification

#contrastive-learning #few-shot #speech

AI research

arXiv cs.AI / cs.LG / cs.CL·11d ago

ECHO: A Matched-Contrast Benchmark for Context-Sensitive Turn-Taking in Full-Duplex Dialogue

#benchmark #chinese #dialogue

AI research

arXiv cs.AI / cs.LG / cs.CL·15d ago

Continue, Adapt, or Yield: In-Turn Adaptation to Overlapping Speech in Full-Duplex Agents

#conversational-ai #evaluation #full-duplex

AI research

arXiv cs.AI / cs.LG / cs.CL·15d ago

MP-Bench: Evaluating Voice Agents as a Multiparty Conversation Participant

#benchmark #conversational-ai #evaluation

AI research

2

The Decoder·16d ago

OpenAI's GPT-Live-1 API lets developers build apps that talk and listen at the same time

#full-duplex #gpt-live-1 #openai

1

OpenAI News·17d ago

Build more natural voice experiences with GPT‑Live‑1 in the API

#api #full-duplex #gpt-live-1

5 min1

Hugging Face Blog·Aug 10, 2026

Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with NVIDIA Magpie TTS

#magpie-tts #nvidia #open-weights

Model release