Glossary → Transcription & speech recognition
Transcription & speech recognition

Real-time factor (RTF)

Real-time factor (RTF) measures transcription speed as processing time divided by audio duration: an RTF of 0.1 means one hour of audio transcribes in six minutes; values below 1 are faster than real time.

Speed claims are often quoted the other way up ('30x real time' means RTF ≈ 0.033). RTF depends on hardware, model size and settings — the same model can be 50x real time on a GPU and slower than real time on a laptop CPU.

For live captioning, RTF must stay well under 1 continuously; for batch transcription, RTF just sets how long you wait.

Related terms

Streaming speech recognitionStreaming speech recognition transcribes audio as it arrives, emitting words within moments of them being spok… Batch transcriptionBatch transcription processes complete, already-recorded audio files, letting the system use full context in b… Quantization (of speech models)Quantization shrinks a neural speech model by storing its weights at lower numeric precision — for example INT…

Frequently asked

What is Real-time factor?

Real-time factor (RTF) measures transcription speed as processing time divided by audio duration: an RTF of 0.1 means one hour of audio transcribes in six minutes; values below 1 are faster than real time.

Why does Real-time factor matter?

Speed claims are often quoted the other way up ('30x real time' means RTF ≈ 0.033). RTF depends on hardware, model size and settings — the same model can be 50x real time on a GPU and slower than real time on a laptop CPU.

What terms are related to Real-time factor?

Closely related concepts: Streaming speech recognition, Batch transcription, Quantization (of speech models) — each has its own entry in this glossary.

Put the term to work

Transcribe audio or video in 99+ languages — speaker labels, word timestamps, captions.

Transcribe a file →