[Bug] Process *Killed* when executing parallel tts commands on different containers (Docker version)

[khaldi-yass]

Describe the bug

Trying to experiment a little bit with running multiple TTS instances at the same time using the docker image, I created 5 different containers, and trying to execute a TTS command on each of those running containers, but only 2 out of 5 produce an actual output, while the others simply log "Killed" and terminate.

Is there a limitation on a shared resource between those 5 separate containers? I though they are running on separate environments.

To Reproduce

1. Download the docker Image.
2. Create 5 different containers with the docker image.
3. Open each container in a different cmd window.
4. On each container past and run this command simultaneously tts --model_name tts_models/multilingual/multi-dataset/xtts_v2 --speaker_idx "Daisy Studious" --language_idx en --text "Hello world." --out_path "out.pcm" --use_cuda true.

Expected behavior

All 5 containers, reproduce 5 outputs separately. Instead 3 gets killed and only 2 produce the outputs.

Logs

> tts_models/multilingual/multi-dataset/xtts_v2 is already downloaded.
 > Using model: xtts
Killed

### Environment

```shell
{
    "CUDA": {
        "GPU": [
            "NVIDIA GeForce RTX 3060"
        ],
        "available": true,
        "version": "11.8"
    },
    "Packages": {
        "PyTorch_debug": false,
        "PyTorch_version": "2.1.1+cu118",
        "TTS": "0.22.0",
        "numpy": "1.22.0"
    },
    "System": {
        "OS": "Linux",
        "architecture": [
            "64bit",
            ""
        ],
        "processor": "x86_64",
        "python": "3.10.12",
        "version": "#1 SMP Fri Mar 29 23:14:13 UTC 2024"
    }
}

Additional context

No response