#speechrecognition

ViVeL — Vida Verdadera De Dios, "True Life of God." I wanted a name shaped like Israel: not just a...

Apple's new on-device SpeechAnalyzer API cut errors roughly fourfold versus its legacy recognizer and outperformed OpenAI's Whisper Small while using about a third of the compute, in an English-only single-vendor benchmark.

Cohere released an open-source (Apache 2.0) Arabic speech-recognition model that it says is the most accurate open Arabic transcriber to date, cutting word error rate well below OpenAI's Whisper Large V3 and handling dialects and Arabic-English code-

AI transcripts often misfire on drug names, dosages, and jargon. Learn to measure accuracy with Missed Entity Rate and the tools that get terms right.

Audio, metadata, and structural context turn generic speaker labels into named, role-assigned participants. See how to configure each one in the API.

Speaker embeddings are the voice "fingerprints" behind diarization. See how the 4-step pipeline labels who spoke when — with code and accuracy benchmarks.

Originally published on andrew.ooo — visit the original for any updates, code snippets that aged...

Global dictation apps falter, but Korea's Naver Clova AI spent a decade perfecting hyper-accurate, multilingual speech-to-text. Discover how Naver voi

In questo progetto ho creato Oreste AI, un assistente vocale italiano basato su JavaScript puro, HTML...

Building a Real-Time AI Voice Agent for Asterisk AudioSocket + Deepgram STT + Groq LLM +...

Tutorial 24: Call Recording Transcription with Faster-Whisper Batch transcription of...

ElevenLabs Cloud Voice Agent with Asterisk SIP Integration Cloud-Hosted Conversational AI...