---
language:
  - et
library_name: k2
tags:
  - automatic-speech-recognition
  - k2
widget:
  - example_title: AK test 1
    src: >-
      https://huggingface.co/TalTechNLP/icefall_pruned_transducer_stateless7_streaming_et/resolve/main/test_wav.wav
license: cc-by-sa-4.0
---

# Icefall streaming ASR model for Estonian

This is a streaming end-to-end transducer model for Estonian, trained using [Icefall](https://github.com/k2-fsa/icefall)

It is trained on around 800 h of manually transcribed speech from various domains and on
about 2500 h of automatically transcribed speech from Estonian TV (mainly news and talkshows)

## Serving

To use it on a server for browser-based ASR:

  * Install [Sherpa](https://github.com/k2-fsa/sherpa)
  * Clone this model locally:
    
      ```
      git lfs install
      git clone https://huggingface.co/TalTechNLP/icefall_pruned_transducer_stateless7_streaming_et
      ```
      
  * Set SHERPA_ROOT_DIR to the sherpa root directory
       
  * Start serving on port 6006:
  
    ```
    sherpa-online-websocket-server --use-gpu=false --decode-chunk-size=32  \
      --encoder-model=icefall_pruned_transducer_stateless7_streaming_et/exp/1d/encoder_jit_trace.pt \
      --decoder-model=icefall_pruned_transducer_stateless7_streaming_et/exp/1d/decoder_jit_trace.pt \
      --joiner-model=icefall_pruned_transducer_stateless7_streaming_et/exp/1d/joiner_jit_trace.pt  \
      --tokens=icefall_pruned_transducer_stateless7_streaming_et/data/lang_bpe_1000/tokens.txt \
      --doc-root=${SHERPA_ROOT_DIR}/sherpa/bin/web --decoding-method=modified_beam_search
    ```
      
  * Open in browser: http://localhost:6006 (also works via ssh tunnel) and go to "Streaming-Record" tab
  
  * Click "Connect" and then "Streaming-Record" button, and start talking