search

FengYen-Chang / DeepSpeech.OpenVINO

This is the demo application for Deep Speech algorithm, which make speech to text that are being performed on input speech audio.
Mozilla Public License 2.0
5 stars 1 forks source link

Some updation of Chinese DeepSpeech #4

Open ShiJiaying opened 1 year ago

ShiJiaying commented 1 year ago

Hi, thanks for you wonderful work, and I have tested the demo code for Chinese DeepSpeech model. There are some things to pay attention to:

  1. Follow the version of OpenVINO as shown below openvino 2023.0.0.dev20230217 openvino-dev 2023.0.0.dev20230217 openvino-telemetry 2022.1.0
  2. ./mo_tf.py is from https://github.com/openvinotoolkit/openvino/blob/master/tools/mo/openvino/tools/mo/mo_tf.py
  3. The L125 in deep_speech_demo.py is https://github.com/FengYen-Chang/DeepSpeech.OpenVINO/blob/5dd81b0148d78e61b7806cc35fd101ec70deaddb/demo/deep_speech_demo.py#L125 Change it to net = ie.read_network(model=model_xml, weights=model_bin)