Build with Voice Harbor
Computing time benchmarks using Voice Harbor Python SDK
Complex sample
Format: mp3
Input Channel: 1
Output Channel: 2
Speaker: 2
Language: English & Spanish
Duration: 5 min 49 sec (11 min 38 sec)
task | Duration | Jon parameters |
---|---|---|
transcribe | Same as 2025-05-07 | "task":"transcribe", "diar":True, "code-switch":True |
phi | Same as 2025-05-07 | "task":"phi", "diar":True, "code-switch":True |
biometric | 8.35 min | "task":"biometric", "diar":True, "code-switch":True |
phi-biometric | 8.51 min | "task":"phi-biometric", "diar":True, "code-switch":True |
Complex sample
Format: mp3
Input Channel: 1
Output Channel: 2
Speaker: 2
Language: English & Spanish
Duration: 5 min 49 sec (11 min 38 sec)
task | Duration | Jon parameters |
---|---|---|
transcribe | 55.71 sec | "task":"transcribe", "diar":True, "code-switch":True |
phi | 85.73 sec | "task":"phi", "diar":True, "code-switch":True |
biometric | 11.96 min | "task":"biometric", "diar":True, "code-switch":True |
phi-biometric | 11.79 min | "task":"phi-biometric", "diar":True, "code-switch":True |
We have released our biometric redaction service, read more about all redaction parameters here Job Parameters. Just ask, and we’ll gladly provide more information on why biometric redaction is the recommended approach to comply 100% with HIPAA and GDPR.
Simple sample
Format: wav
Input Channel: 1
Output Channel: 1
Speaker: 1
Language: English
Duration: 0:51 sec
task | Duration | Jon parameters |
---|---|---|
transcribe | 14.12 sec | "task":"transcribe" |
phi | 24.10 sec | "task":"phi", "model":"mini" |
biometric | 75.12 sec | "task":"biometric" |
phi-biometric | 75.09 sec | "task":"phi-biometric" |