wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #4522 from pipecat-ai/mb/stt-latency-smallest

Browse Source 
Add P99 latency for Smallest AI, Mistral, XAI STT
This commit is contained in:
Mark Backman
2026-05-20 11:21:00 -04:00
committed by GitHub
parent 4a96ab7073 38407e091d
commit be93350eae
2 changed files with 4 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
changelog/4522.changed.md Normal file
View File

@@ -0,0 +1 @@
- Updated the default p99 TTFS latency values for Smallest AI, Mistral, and XAI STT so turn stop timing uses measured values instead of the conservative fallback.

8
src/pipecat/services/stt_latency.py
View File

@@ -44,17 +44,15 @@ GLADIA_TTFS_P99: float = 1.49
GOOGLE_TTFS_P99: float = 1.57 GOOGLE_TTFS_P99: float = 1.57
GRADIUM_TTFS_P99: float = 1.61 GRADIUM_TTFS_P99: float = 1.61
GROQ_TTFS_P99: float = 1.54 GROQ_TTFS_P99: float = 1.54
MISTRAL_TTFS_P99: float = 1.89
OPENAI_TTFS_P99: float = 2.01 OPENAI_TTFS_P99: float = 2.01
OPENAI_REALTIME_TTFS_P99: float = 1.66 OPENAI_REALTIME_TTFS_P99: float = 1.66
SARVAM_TTFS_P99: float = 1.17 SARVAM_TTFS_P99: float = 1.17
SMALLEST_TTFS_P99: float = 1.59
SONIOX_TTFS_P99: float = 0.35 SONIOX_TTFS_P99: float = 0.35
SPEECHMATICS_TTFS_P99: float = 0.74 SPEECHMATICS_TTFS_P99: float = 0.74
XAI_TTFS_P99: float = 2.14
# These services run locally and should be replaced with measured values # These services run locally and should be replaced with measured values
NVIDIA_TTFS_P99: float = DEFAULT_TTFS_P99 NVIDIA_TTFS_P99: float = DEFAULT_TTFS_P99
WHISPER_TTFS_P99: float = DEFAULT_TTFS_P99 WHISPER_TTFS_P99: float = DEFAULT_TTFS_P99
# No benchmark available yet; using conservative default
MISTRAL_TTFS_P99: float = DEFAULT_TTFS_P99
SMALLEST_TTFS_P99: float = DEFAULT_TTFS_P99
XAI_TTFS_P99: float = DEFAULT_TTFS_P99