Other
Speech to Text Benchmarking Methodology (Artificial Analysis)
Artificial Analysis, Artificial Analysis, n.d.
Read the originalRead on October 17, 2026
- What it is
- Artificial Analysis's written method for its speech to text benchmarks, including the speed factor and the two streaming latency metrics.
- How much weight it carries
- The method the benchmark publishes, so it is the basis for reading its numbers. It is a benchmark's own description, not an external standard.
Facts taken from this source
- Artificial Analysis says a speed factor above 1 means the service transcribes faster than real time, so 2.0 means 10 minutes of audio is transcribed in 5 minutes.
“A value above 1 means the service transcribes faster than real time” (Speed Factor (batch))
- Artificial Analysis defines time to first partial transcript as the time of the first partial transcript shared after the Silero voice activity detector detects end of speech.
“Time to First Partial Transcript: time of first Partial-denoted transcript shared after SileroVAD detected end of speech.” (Latency Metrics (streaming))