assemblyai-stt-3-7-realtime
Streaming model for notetakers, captions and agent assist (launched Nov 17, 2026) with realtime speaker labels across 32 languages; not tuned for agent turn-taking. Replaces Universal-Streaming English and Multilingual.
01 Price
$0.30per hour
Cheaper than 72% of 238 priced speech-to-text and realtime models.
Speech-to-text and realtime $ per hour of audio, log scale 238 priced
other models this modelAll models by price
Billed per hour of session duration (WebSocket open time, not audio sent). Keyterms prompting included. Add-ons (/hr): streaming speaker diarization 0.12, general prompting (beta) 0.05, medical or vet mode 0.15, PII text redaction 0.12.
Pricing source, checked 2026-11-19
02 Languages 32
03 Features 4 of 11 listed
- Word timestampsnot listed
- Speaker diarization
- Punctuationnot listed
- Profanity filternot listed
- Custom vocabulary
- Language detection
- Translationnot listed
- Sentimentnot listed
- PII redaction
- Summarizationnot listed
- Topic detectionnot listed
"Not listed" means the provider's data doesn't mention it, not that the model lacks it.
04 Get this record
curl https://voicedex.dev/v1/providers/assemblyai/models/assemblyai-stt-3-7-realtime
Free, no key needed. Use model id assemblyai-stt-3-7-realtime with AssemblyAI's own API. API reference.