Machine Learning2 min read
Jack Dorsey’s Block made an AI model that talks to you like an awkward new friend
Covered by 5 sources
Updates on Speech Recognition
Every AI story we track on Speech Recognition: 44 stories so far, each summarized in our own words and linked to the publisher that reported it.
Pulled from 125 sources
This week
Machine Learning5 min read
Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster
Covered by 2 sources
Machine Learning46 min read
A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU
MarkTechPost
This month
Machine Learning4 min read
Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint
Covered by 2 sources
Agents4 min read
Grok Voice 3.0 from SpaceXSI outscores OpenAI and Google on speech-to-speech index
Covered by 4 sources
Machine Learning2 min read
Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models
Apple Research Blog
Machine Learning5 min read
TII Releases Falcon-ASR: 1.6B Arabic Speech Model Hits 20.92% WER
MarkTechPost
Business & Enterprise4 min read
Zoom CTO XD Huang on bringing Steve Jobs to every meeting
The Next Web
Machine Learning4 min read
Microsoft AI Releases MAI-Transcribe-2-Streaming: #1 Real-Time Speech-to-Text Model on Artificial Analysis
MarkTechPost
Machine Learning5 min read
Microsoft targets ultra-realistic voice agents with its first streaming transcription model
Covered by 3 sources
Machine Learning6 min read
Open TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning
Hugging Face
Machine Learning7 min read
SuperWhisper s1-mini: The 600M Parameter Model Built Just for Transcription
KDnuggets
LLM & Text Generation5 min read
Alibaba Qwen Releases Qwen-Audio-3.1-Realtime: A Full-Duplex Voice Model Trained to Think, Act, and Decide When to Speak
MarkTechPost
Machine Learning5 min read
Sarvam AI Releases Saaras V4: A Speech-to-Text Model for All 22 Indian Languages and Global English
MarkTechPost
Machine Learning16 min read
**Know Who Spoke When: Build Real-Time, Multi-Speaker AI with NVIDIA Nemotron 3 Diarization**
Covered by 2 sources
Machine Learning2 min read
Compressing Streaming Neural Audio Encoders via Latent-Space Distillation
Apple Research Blog
Machine Learning4 min read
Transcription now costs 10 cents an hour. The differentiator is no longer the model, it is whose conversations you get to train on.
Covered by 3 sources
Machine Learning6 min read
GPT-6 Astra 🚀, Grok Bot Enterprise 🤖, Runway world model 🌎
Covered by 2 sources
LLM & Text Generation3 min read
Meta's new AI transcription model can distinguish between multiple speakers and languages in real-time
Covered by 4 sources
Education4 min read
Verba: Learn a language by talking to an AI never touches the internet
Hacker News
Machine Learning14 min read
The Open ASR Leaderboard Adds Its First Global South Language
Hugging Face
Machine Learning10 min read
Deepgram deepens Amazon SageMaker AI observability with Enhanced Metrics
AWS Blog
Machine Learning5 min read
IBM Says Granite Speech 5.0 Transcribes 3.5 Hours of Speech in One Second
Unite.AI
LLM & Text Generation6 min read
Wispr raised $280mn at $2bn to argue the text box is finished
Covered by 2 sources
LLM & Text Generation2 min read
Wispr raises $280M at $2B valuation as it looks beyond dictation
TechCrunch
Agents4 min read
PolyAI Releases Dialog-RSN-1: An Audio-Native Dialog Model That Fuses Turn-Taking, Speech Recognition, Function Calling, And Response
MarkTechPost
Run NVIDIA Nemotron 3 Nano as a fully managed serverless model on Amazon Bedrock
AWS Blog
That's everything we have on Speech Recognition right now