Speech Recognition

5 articles in this category

AI NewsSpeech RecognitionNatural Language Processing

Revolutionizing Voice AI: Deepgram's Quest for Universal Reliability

Deepgram improves speech-to-text accuracy by 30%

Feb 13, 2026

AI NewsSpeech RecognitionLow-Resource Languages

Elevating Voices in AI: Microsoft Research Launches Paza & PazaBench

Microsoft Research unveils Paza, a human-centered speech pipeline, and PazaBench, the first leaderboard for low-resource languages, covering 39 African languages and 52 models.

Feb 5, 2026

AI NewsSpeech RecognitionAI Agents

Microsoft Releases VibeVoice-ASR: A Unified Speech-to-Text Model for Long-Form Audio

Microsoft’s VibeVoice-ASR tackles long-form audio transcription, achieving 60-minute single-pass processing with structured output.

Jan 22, 2026

AI NewsSpeech RecognitionNVIDIA

NVIDIA Releases Nemotron Speech ASR: Low-Latency Speech Recognition

NVIDIA released Nemotron Speech ASR, an open-source transcription model achieving approximately 7.84% WER at a 0.16s chunk size for low-latency applications.

Jan 6, 2026

AI NewsGemini ModelsSpeech Recognition

Improved Gemini audio models for powerful voice interactions

Google’s upgraded Gemini 2.5 Native Audio model achieves a 71.5% score on ComplexFuncBench Audio, improving voice agent capabilities.

Dec 12, 2025