nvidia
/

stt_ka_fastconformer_hybrid_transducer_ctc_large_streaming_80ms_pc

Automatic Speech Recognition

Georgian language

Model card Files Files and versions Community

SKostandian commited on Dec 11, 2024

Commit

56b5919

·

verified ·

1 Parent(s): f218fdc

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -58,7 +58,7 @@ model-index:
 This collection contains large size versions of cache-aware FastConformer-Hybrid (around 115M parameters) trained on a Georgian speech.
 These models are trained for streaming ASR with look-ahead of 80ms which be used for very low-latency streaming applications.
-All models are hybrid with both Transducer and CTC decoders.
 ## Model Architecture

 This collection contains large size versions of cache-aware FastConformer-Hybrid (around 115M parameters) trained on a Georgian speech.
 These models are trained for streaming ASR with look-ahead of 80ms which be used for very low-latency streaming applications.
+The model is hybrid with both Transducer and CTC decoders.
 ## Model Architecture