Real-time STT Platform

リアルタイム音声認識

Makima provides real-time speech-to-text transcription with speaker diarization. Stream audio from your microphone or upload files for instant transcription.

高精度なリアルタイム音声認識エンジン。WebSocket 経由でストリーミング処理を行い、話者分離にも対応。

Start Listening GitHub soryu.co

FEATURES//MAKIMA

Platform Features

Real-time Streaming
WebSocket-based audio streaming with low latency transcription

Speaker Diarization
Automatic speaker identification and labeling

End-of-Utterance
Smart detection of speech boundaries

Multi-format
Support for PCM32F and PCM16 audio encoding

API // makima.jp endpoint: wss://api.makima.jp

{`[001] endpoint ............... /api/v1/listen
[002] protocol ............... WebSocket
[003] encoding ............... pcm32f / pcm16
[004] sample_rate ............ 16000 Hz (default)
[005] features ............... transcription, diarization, eou`}