README.md: add instructions for running client
Specify command to run client script. Signed-off-by: Jeny Sadadia <jeny.sadadia@collabora.com>
This commit is contained in:
@@ -103,7 +103,15 @@ If you don't want this, set `--no_single_model`.
|
||||
|
||||
|
||||
### Running the Client
|
||||
- Initializing the client with below parameters:
|
||||
|
||||
Use the below command to run the client:
|
||||
```bash
|
||||
python3 run_client.py --files <audio-file-name>
|
||||
```
|
||||
This will connect to the localhost server running on port 9090 by default. Use flags `--server` and `--port` to use different configurations. The above command will transcribe audio file provided with `--files` flag.
|
||||
|
||||
|
||||
Here are the details of client instance implemented in `run_client.py` script:
|
||||
- `lang`: Language of the input audio, applicable only if using a multilingual model.
|
||||
- `translate`: If set to `True` then translate from any language to `en`.
|
||||
- `model`: Whisper model size.
|
||||
|
||||
Reference in New Issue
Block a user