model details

src/main.py

@@ -156,7 +156,9 @@ async def model_command(self, update: Update, context: CallbackContext) -> None:
  "- <b>Large</b>: Most precise, processes at real-time (1x speed), uses ~10GB VRAM.\n\n"
  "Note: '.en' models (e.g., 'tiny.en') are optimized for English and offer better accuracy for English audio. "
  "As model size increases, the benefit of English optimization becomes less significant. Choose based on your "
- "needs for speed, memory usage, and linguistic accuracy.",
+ "needs for speed, memory usage, and linguistic accuracy. "
+ "As a general guideline, larger models are more accurate but also slower. "
+ "I.e. best balance in speed vs. accuracy in English language is likely <code>medium.en</code>.",
  parse_mode='HTML')
  return