Clarify training hardware (A100, ~13 hours) vs inference (any GPU/CPU)

README.md

@@ -17,10 +17,11 @@ pipeline_tag: text-to-speech
 A from-scratch Burmese (Myanmar) text-to-speech model. 31M parameters, ~63 MB (fp16).
 
 - **Language**: Burmese (မြန်မာဘာသာ)
-- **Training**: ~1.34M samples (news + real-world audio)
-- **Steps**: 81,000
+- **Training data**: ~1.34M samples (news + real-world audio)
+- **Training steps**: 81,000
+- **Training hardware**: A100-80GB (Colab Pro+), ~13 hours, ~88 compute units
+- **Inference hardware**: Free Colab T4, any modern GPU, or CPU (slower)
 - **Architecture**: StableTTS (DiT + flow matching) + Vocos vocoder
-- **Runs on**: Free Colab T4, CPU (slower), any modern GPU
 - **License**: **CC0 1.0** (public domain, no attribution required)
 
 ---