How Voice AI Companies Use Live Events to Build Trust and Shape Buyer Thinking
Voice AI companies that consistently host and actively participate in professional events gain early visibility...
Speech & Voice
Latest speech & voice
Innovations in AI speech and AI voice, including live translation, voice cloning, speech-to-speech models, and speech-to-text
Array
(
[post_type] => post
[posts_per_page] => 12
[paged] => 1
[post_status] => publish
[tax_query] => Array
(
[relation] => AND
[0] => Array
(
[taxonomy] => category
[field] => term_id
[terms] => 39439
[operator] => IN
)
)
)
Breakfasts, Game Nights, Car Rides: The PECII Corpus for Natural Language Interactions
Researchers from three universities collected a wealth of natural human interactions across four languages and...
OpenAI Updates Realtime Model to Improve Multilingual Voice Agent Reliability
OpenAI updates its gpt-realtime voice model to improve reliability in multilingual voice agents for customer...
Prompt-Based Control Reaches Enterprise Speech-to-Text
Recent releases from Microsoft and AssemblyAI reflect growing interest in structured, configurable speech recognition as...
What Text-to-Speech Evaluation Misses in Real-World Deployment
Researchers argue that text-to-speech evaluation under-tests what matters in real-world deployment — and propose a...
Microsoft Unveils VibeVoice-ASR for Long-Form, Multi-Speaker Transcription
Microsoft releases VibeVoice-ASR, an open-source speech-to-text model designed for long-form audio, structured transcription, and customised...
Alibaba Expands Speech Stack with New Open-Source Models
Alibaba’s Qwen team releases a new set of open-source speech models under the Qwen3 family,...
Researchers Begin Benchmarking Cross-Lingual Voice Cloning
A new cross-lingual voice cloning track at IWSLT 2026 highlights changing priorities in multilingual speech...
NVIDIA Doubles Down on Open Speech AI
NVIDIA doubles down on open speech AI with ultra-low-latency automatic speech recognition and multilingual text-to-speech...
Google Launches MedASR, an Open Medical Speech-to-Text Model
Google introduces MedASR, an open-weight medical speech-to-text model positioned as a foundational layer for healthcare...
Resemble AI Open-Sources Chatterbox Turbo
Resemble AI releases an open-source text-to-speech model designed for real-time, expressive voice generation and positioned...