Picute

Japanese audio and video to text

Paste a YouTube link or upload a recording in Japanese. Picute writes out what was said, with a timestamp on every word.

Enter YouTube URL
0/10 URLs
You can enter up to 10 YouTube URLs.

The basics

Input
YouTube links (up to 10 at a time), video or audio files, Google Drive
Languages
Speech recognition in 60 languages
Output
SRT and VTT subtitle files, or an MP4 with the subtitles burned in
Length and size
No limit on video length or file size
Cost
1 credit per minute of transcription. Sign-up includes 10 credits, and free-plan exports carry a watermark. See pricing

Set the spoken language

Choose Japanese as the spoken language in the form, or leave it on auto-detect.

About Japanese (日本語)

Speakers
Around 125 million speakers
Writing system
A mix of kanji, hiragana and katakana, written without spaces between words
Language family
Japonic
Varieties
Standard (Tokyo) Japanese plus regional dialects such as Kansai-ben

What makes Japanese hard to transcribe

Japanese is written without spaces and is full of homophones that only the correct kanji can tell apart, so segmenting words and choosing characters is the core challenge.

Other languages Picute transcribes