Modulate Transcription API

#28 in Transcription (STT)

Modulate · 7× · last seen Jun 30, 2026

1

Momentum

The Modulate Transcription API (product name "Velma Transcribe") is a Speech-to-Text API from voice AI company Modulate, optimized for accuracy on real-world, messy conversation recordings (overlaps, background noise, accents) rather than studio recordings. It supports both batch and real-time streaming transcription via REST and WebSocket, with over 50 supported languages, speaker diarization, timestamps, and optional PII/PHI redaction. The API is billed on a usage basis, starting at very low hourly rates, and positions itself as the most cost-effective and, according to the manufacturer, most accurate solution compared to providers like Deepgram, AssemblyAI, ElevenLabs, or OpenAI.

Sources (7)