Search Articles — Sudonull

Search Results

In this project

Bot Architecture with Whisper.cpp and aiogram 3.x

https://sudonull.com/bot-architecture-with-whisper-cpp-and-aiogram-3-x

Breakdown of asynchronous integration of Whisper.cpp into Telegram bot for Spanish pronunciation assessment. FSM, word-by-word scoring, code examples. For middle/senior dev.

Whisper Speech Recognition in C# for Meetings

https://sudonull.com/whisper-speech-recognition-in-c-for-meetings

Implement RT STT and audio translation from Teams: Whisper Small, Azure/OpenAI. VAD settings, statistics. For C# developers — code, benchmarks, integration. Study it.

Voice Input Claude Russian Offline

https://sudonull.com/voice-input-claude-russian-offline

Set up local Whisper for dictation in Claude.ai: Flask server, rumps Menu Bar, Chrome MV3. Offline, free, no limits. Instructions for Mac with code.

Voice Input for Developers 2026: Comparison of Solutions

https://sudonull.com/voice-input-for-developers-2026-comparison-of-solutions

Testing Whisper Large, Handy and GigaAM v3 for Russian-English dictation in LLM. Selection criteria, benchmarks, setup for technical tasks. Learn how to speed up workflow.

Microsoft MAI models: WER 3.8% and plans until 2027

https://sudonull.com/microsoft-mai-models-wer-3-8-and-plans-until-2027

Analysis of new Microsoft models: MAI-Transcribe-1 outperforms Whisper and Gemini in transcription accuracy. Plans for frontier AI by 2027. For developers — benchmarks, comparisons, and infrastructure. Explore the details.

From other projects

FDA approved the first gene therapy for hereditary deafness

https://ymaho.com/fda-approved-the-first-gene-therapy-for-hereditary-deafness

Regeneron's Otarmeni restores hearing to normal in 42% of patients. The therapy treats OTOF gene mutations with a single injection. Learn how AAV vector technology has for the first time restored the ability to hear a whisper.

From the web

Whisper AI: Official Whisper Transcription | Powered by OpenAI

https://whisperai.com/

Transcribe audio and video with OpenAI Whisper accuracy — unlimited AI transcription, 100+ languages, and speaker labels. Free to start, no credit card required.

Introducing Whisper - OpenAI

https://openai.com/index/whisper/

Sep 21, 2022 · Whisper is an automatic speech recognition (ASR) system trained on 680,000 hours of multilingual and multitask supervised data collected from the web. We show that the use of such a …

GitHub - openai/whisper: Robust Speech Recognition via Large-Scale …

https://github.com/openai/whisper

Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multitasking model that can perform multilingual speech recognition, speech translation, …

Wispr Flow | Effortless Voice Dictation

https://wisprflow.ai/

Flow makes writing quick and clear with seamless voice dictation. It is the fastest, smartest way to type with your voice.

Download Whisper App for Mac, Windows, Linux — Free, No Signup

https://www.app-whisper.com/download

Whisper App stays invisible during screen sharing, transcribes your meetings on-device, and runs live AI coaching during the call — without a meeting bot. Your audio never leaves your machine.

Wir stellen vor: Whisper - OpenAI

https://openai.com/de-DE/index/whisper/

Sep 21, 2022 · Wir haben ein neuronales Netz namens Whisper trainiert und stellen es als Open Source zur Verfügung. Whisper ist bei der Erkennung englischer Sprache fast so robust und genau ist wie …

Was ist OpenAI Whisper? So nutzt man es - CHIP

https://www.chip.de/ratgeber/kuenstliche-intelligenz/was-ist-openai-whisper-so-nutzt-man-es_e901140a-026f-4fe7-95ea-a271190fa52f.html

Dec 17, 2024 · Whisper von OpenAI ist ein Spracherkennungssystem, mit dem Sie gesprochene Texte transkribieren und übersetzen lassen können. Das Programm beherrscht viele Sprachen.

Whisper Web — Free AI Speech Recognition | Browser-Based

https://whisperweb.app/

Whisper Web processes everything directly in your browser using WebAssembly. Your audio files never leave your device, never get uploaded to any servers, and are never stored anywhere.

Fortschrittlichste KI-Transkriptionsplattform | Sprache zu Text | WhisperAI

https://whisperai.com/de

WhisperAI transkribiert Audio mit 99 % Genauigkeit mithilfe von OpenAI Whisper. Unbegrenzte KI-Transkription in über 100 Sprachen mit Sprechererkennung.

openai/whisper-large-v3 · Hugging Face

https://huggingface.co/openai/whisper-large-v3

Dec 6, 2022 · We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Trending Now