ブログ

音声、プライバシー、
音声入力とAIノート。

オンデバイス文字起こしの仕組み、モデルの選び方、そして録音をもっと活用する方法についての解説記事。

このページは英語で書かれています。

記事

Cloud transcription API pricing, per hour of audio

Cloud transcription API pricing per hour of audio: verified batch and streaming list prices, worked annual costs, and the charges not in the rate.

読了目安 12 分
記事

How much RAM a Mac needs for Whisper and a local LLM

How much RAM a Mac needs for Whisper, Parakeet and a local LLM: unified memory, model sizes by quantization, KV cache growth, and 8, 16, 32 and 64 GB budgets.

読了目安 12 分
記事

How Speaker Diarization and Speaker Recognition Work

Speaker diarization explained in plain words: how software splits audio by speaker, matches voices across recordings, where it fails, and how to help it.

読了目安 8 分
記事

Live captions and transcripts as an accessibility tool

How live captions and recorded transcripts solve different problems for deaf and hard of hearing people, where each one helps, and what this kind of app is not.

読了目安 9 分
記事

Offline speech to text on iPhone: what the platform allows

Offline speech recognition on iPhone: Apple's on-device dictation, the iOS 26 SpeechAnalyzer API, Whisper models, and why keyboard extensions cannot use the mic.

読了目安 9 分
記事

On-device transcription vs cloud: privacy, cost and speed

What on-device transcription means technically, what cloud services do with your audio, the per-minute billing math, and the cases where cloud still wins.

読了目安 8 分
記事

Private Medical and Legal Dictation on Mac and iPhone

Why clinicians and lawyers need a private medical dictation app or legal dictation software that keeps audio on the device, and what to check with compliance.

読了目安 9 分
記事

Whisper vs Parakeet: Which On-Device Speech Model Is Better?

Whisper vs Parakeet TDT compared in plain words: architecture, speed, languages, accuracy, hallucinations and when to pick which on-device speech to text model.

読了目安 9 分
記事

Why Whisper hallucinates and repeats itself, and how to stop it

Whisper hallucination and repetition loops explained: why silence becomes 'Thank you for watching' and the whisper.cpp, openai-whisper and faster-whisper fixes.

読了目安 10 分