Update README.md

README.md

@@ -19,6 +19,10 @@ pipeline_tag: text-to-speech
 
 Preview-release for Fosdem 2025 with current training epochs (Training is still ongoing)
 
+Demos: https://huggingface.co/spaces/Banafo/Kroko-Streaming-ASR-Wasm (running on CPU in the browser).
+Gradio / python: https://huggingface.co/spaces/Banafo/Kroko-Streaming-ASR-Python
+
+
 This is family of low-latency streaming models for use on edge devices.
 Goals:
 - Faster and or higher quality than similar sized Whisper and other models.
@@ -27,7 +31,9 @@ The license is still under consideration (probably Coqui), it will be dual licen
 - free for non commercial use
 - Affordable license for commercial use.
 
-English, French, German available. More languages in preparation.
+English, French, German available. Spanish and Portuguese by 14 feb.
 
+Training is done with modified k2/ Icefall. 
 Inference can be done with standard Sherpa project. 
+
 Big thanks and shoutout to the lhotse / Sherpa / k2 / Icefall team.