Send initial file with duration limit (#830)

<!-- This is an auto-generated comment: release notes by OSS
Entelligence.AI -->
### Summary by Entelligence.AI

- New Feature: Updated the transcription functionality to include a
duration limit. This ensures that data transmission stops when the total
time exceeds the specified duration, enhancing efficiency and control
over resource usage.
- Refactor: Adjusted the calculation of transcription duration based on
the length of the speech profile for improved accuracy.
- Bug Fix: Modified the duration calculation logic to prevent potential
overestimation of transcription time.
<!-- end of auto-generated comment: release notes by OSS Entelligence.AI
-->

backend/routers/transcribe.py

@@ -180,11 +180,12 @@ def stream_audio(audio_buffer):
     duration = 0
     try:
         if language == 'en' and codec == 'opus' and include_speech_profile:
+            second_per_frame: float = 0.01
             speech_profile = get_user_speech_profile(uid)
-            duration = get_user_speech_profile_duration(uid)
+            duration = len(speech_profile) * second_per_frame
             print('speech_profile', len(speech_profile), duration)
             if duration:
-                duration *= 2
+                duration += 10
         else:
             speech_profile, duration = [], 0