Mastodawn

Petal für #macOS: Simple Diktat-App für die Menüleiste mit #Qwen3, #Parakeet und #Voxtral. #OpenSource, ohne Cloud und Abo 🎙️🖥️🔓 https://tchgdns.de/?p=162724

sayzard Mar 13

Nico Martin (@nic_o_martin)

MistralAI의 Voxtral과 Transformers.js, WebGPU 조합으로 브라우저에서 실시간 음성 전사가 가능해졌다는 발표입니다. 다양한 언어를 지원하며 문장 중간에 언어가 바뀌어도 인식하는 기능을 강조하여 웹 기반 ASR(자동 음성인식)의 저지연·다국어 적용 사례로 의미가 큽니다.

https://x.com/nic_o_martin/status/2032087412462022663

#mistralai #voxtral #transformersjs #webgpu #speechrecognition

🤷 Nico Martin (@nic_o_martin) on X

Voxtral (by @MistralAI) + Transformers.js + WebGPU enables real-time speech transcription in the browser. It also supports a wide range of languages, which are even recognized in the middle of a sentence 🚀

X (formerly Twitter)

Michael Walder Feb 10

🚀 Sprachenlernen 2.0: Wenn KI auf Vokabelheft trifft.
Wir haben unseren Vokabeltrainer bei #didaquiz komplett überarbeitet. Das Ziel: Weg vom sturen Auswendiglernen, hin zu echtem Spracherwerb.

Das Herzstück ist der neue Sprech-Modus: Wir integrieren #Voxtral von #mistralai um die Aussprache der Kinder präzise zu analysieren und direktes Feedback zu geben. 🗣️✅

Zusätzlich bietet das Tool Modi zum Erkennen und Schreiben – alles nahtlos integriert in unser Unified UI. Und weil uns Datenhoheit wichtig ist, basiert alles auf dem offenen #Bitmark Standard.

Teste es selbst auf: https://didaquiz.ch/plugins/exampapers/ep-public.php?ep_id=151

#FediLZ #EdTech #OpenStandards #AI #Education #Innovation #OER #LanguageLearning #didaquiz

Vocabulaire 5.3

Mit didaquiz spielend lernen: Übe den Wortschatz aus dem Magazine 5.3

sayzard Feb 9

Victor M (@victormustar)

Voxtral-Subtitles라는 앱이 Hugging Face Spaces에 공개되어 비디오를 단어 단위 자막으로 전사하고 화자 분리(speaker diarization)와 다국어 번역 기능을 제공한다는 소개. 개발자들이 Spaces에서 바로 시도해볼 수 있는 자막·음성-텍스트 변환 툴임.

https://x.com/victormustar/status/2020776196343238656

#voxtral #subtitles #huggingface #speechtotext #diarization

Victor M (@victormustar) on X

Great app: Voxtral-Subtitles to transcribe any video with word-level subtitles, speaker diarization & multilingual translation. ⬇️ Try in now on Hugging Face Spaces

X (formerly Twitter)

Andreas Becker Feb 6

Mistral veröffentlicht Voxtral Realtime und Mini Transcribe V2. Die Modelle erreichen eine Latenz von unter 200 Millisekunden und sind für den lokalen Einsatz optimiert. Mit einem Transkriptionspreis von 0,003 US-Dollar pro Minute zielt die Strategie auf Unabhängigkeit von US-Cloud-Anbietern und ermöglicht datenschutzkonforme On-Device-Lösungen.

#MistralAI #Voxtral #OpenAI
https://www.all-ai.de/news/news26/mistral-voxtral-v2-sprache-ki

Mistral Voxtral V2 fordert OpenAI mit extrem niedrigen Preisen heraus

Das neue Sprachmodell bietet Echtzeit-Transkription unter 200 Millisekunden und setzt auf lokale Verarbeitung für maximale Datensicherheit.

All-AI.de

sayzard Feb 5

antirez (@antirez)

MistralAI가 실시간으로 동작하는 오픈 가중치 전사(음성→텍스트) 모델 'Voxtral Mini 4B'를 공개했다는 소식입니다. 이어 Whisper.cpp 사례를 따라 C언어 기반 추론 파이프라인(라이브러리 형태)도 준비되었다고 안내하고 있어, 실사용·임베디드 환경에서의 음성 전사 활용과 오픈 소스 배포 측면에서 의미가 큽니다.

https://x.com/antirez/status/2019516583811674148

#mistralai #speechtotext #realtime #opensource #voxtral

antirez (@antirez) on X

Yesterday @MistralAI released an open weights transcription model able to work in real time, Voxtral Mini 4B. Today, following the Whisper.cpp lesson, here is a C inference pipeline ready to use as a library, I hope you'll enjoy it: https://t.co/O7pS228DME

X (formerly Twitter)

Erik Jonker Feb 5

Great new audio transcription model by Mistral.
Voxtral transcribes at the speed of sound. https://mistral.ai/news/voxtral-transcribe-2
#mistral #ai #europe #voxtral #audiotranscription

Voxtral transcribes at the speed of sound. | Mistral AI

Precision diarization, real-time transcription, and a new audio playground.

sayzard Feb 4

Simon Willison (@simonw)

Hugging Face 스페이스의 mistralai/Voxtral-Mini-Realtime 데모 추천: 'No microphone found' 메시지는 무시하고 'Record'를 눌러 브라우저에서 마이크 권한을 허용하면 해결되며, 거의 실시간으로 매우 정확한 음성 전사를 보여준다는 평가.

https://x.com/simonw/status/2019116012969554214

#huggingface #speechrecognition #realtime #mistral #voxtral

Simon Willison (@simonw) on X

The demo on https://t.co/ozhOG8iPeH is worth a try - ignore the "No microphone found" message, clicking "Record" and allowing your browser to use a microphone fixes that. It transcribes very accurately in almost real-time. It's really impressive.

X (formerly Twitter)

TechGlimmer Feb 4

Voxtral Transcribe 2 from Mistral AI brings open, production ready speech AI to everyone: fast, accurate transcription, solid diarization and support for long, multilingual audio. It is a strong option if you want powerful speech understanding without locking into closed APIs.

#Voxtral #Transcribe2 #MistralAI #SpeechAI #AITranscription #OpenSourceAI #FLOSS