https://github.com/rhasspy/bemused-client
Streaming TFLite keyword detector
https://github.com/rhasspy/bemused-client
Last synced: about 2 months ago
JSON representation
Streaming TFLite keyword detector
- Host: GitHub
- URL: https://github.com/rhasspy/bemused-client
- Owner: rhasspy
- License: mit
- Created: 2021-03-04T16:59:02.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2021-06-16T13:14:54.000Z (almost 4 years ago)
- Last Synced: 2025-01-16T22:43:06.884Z (3 months ago)
- Language: Python
- Size: 14.6 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Bemused Client
Streaming TFLite-based keyword detector.
New models can be trained with https://github.com/google-research/google-research/tree/master/kws_streaming
## Dependencies
* Python 3.7 or higher
* Tensorflow 2.4
* sounddevice
## Model Format
Models should be placed in a directory with:
* model.tflite - the streaming TFLite model file
* config.json - JSON configuration for bemused
### Model Configuration
Below is a sample configuration for a "raspberry" keyword:
```json
{
"sample_rate": 16000,
"num_channels": 1,
"block_ms": 20,
"keyword_labels": [
"raspberry"
],
"all_labels": [
"_silence_",
"_unknown_",
"raspberry",
"_not_kw_"
]
}
```
## Usage
```sh
$ python3 -m bemused_client /path/to/model/directory
```
See `bemused_client --help` for more options