Articles by tag: speech-recognition

Speech Recognition in MAX: Server Models

Learn how MAX uses dynamic ML models for call analysis. We explain the architecture, risks, and benefits. Read the privacy analysis and updates without app updates.

Qwen3.5-Omni: AI generates code from video

Qwen3.5-Omni from Alibaba: multimodal AI with 256K context, SOTA in audio-video. Generates code from screen recordings. Integrate into projects — read details and benchmarks.

Speech Recognition: 80% Accuracy Limit

Learn why speech recognition systems haven't exceeded 80% accuracy for 25 years. Semantics, grammar, and context as barriers. For developers: analysis of problems and solutions. Read the details.

NPU on MCU for speech: STM32N6 and acoustics

Acoustic speech recognition model on STM32N6 with NPU: PER 5.3%, 0.2 W, real time. Comparison, metrics, application for developers. Study the implementation.

C# speech recognition VK Cloud with hooks

Implementation of a voice typer in C# with NAudio, low-level hooks and VK Cloud API. Full code, error handling, STA for Clipboard. For middle/senior dev — assemble and test.

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.