LocalAI Parler-tts doesn't work when installed from gallery, documentation unhelpful

LocalAI version: v2.16.0-cublas-cuda12-ffmpeg

Environment, CPU architecture, OS, and Version: Running the v2.16.0-cublas-cuda12-ffmpeg image in Kubernetes.

Describe the bug When I install Parler using the webui gallery and try to run it, I only get an error that the model could not be found. It doesn't seem to download a model anywhere. The documentation on the LocalAI website says it can be installed and configured through the gallery, but installing it through that doesn't download a model, I can't find any way to configure it through the webui, and I can't find anything anywhere describing how to configure it manually.

To Reproduce Open the webui, go to the Models tab, find parler-tts-mini-v0.1 and install it. Go to the TTS tab, load parler-tts-mini-v0.1, and try to prompt it.

Expected behavior I get an output like with Piper.

Logs

Debug Output

11:41PM INF Loading model with backend parler-tts 11:41PM DBG Stopping all backends except '' 11:41PM DBG Loading model in memory from file: /models 11:41PM DBG Loading Model with gRPC (file: /models) (backend: parler-tts): {backendString:parler-tts model: threads:0 assetDir:/tmp/localai/backend_data context:{emptyCtx:{}} gRPCOptions:0xc0004d6248 externalBackends:map[autogptq:/build/backend/python/autogptq/run.sh bark:/build/backend/python/bark/run.sh coqui:/build/backend/python/coqui/run.sh diffusers:/build/backend/python/diffusers/run.sh exllama:/build/backend/python/exllama/run.sh exllama2:/build/backend/python/exllama2/run.sh huggingface-embeddings:/build/backend/python/sentencetransformers/run.sh mamba:/build/backend/python/mamba/run.sh openvoice:/build/backend/python/openvoice/run.sh parler-tts:/build/backend/python/parler-tts/run.sh petals:/build/backend/python/petals/run.sh rerankers:/build/backend/python/rerankers/run.sh sentencetransformers:/build/backend/python/sentencetransformers/run.sh transformers:/build/backend/python/transformers/run.sh transformers-musicgen:/build/backend/python/transformers-musicgen/run.sh vall-e-x:/build/backend/python/vall-e-x/run.sh vllm:/build/backend/python/vllm/run.sh] grpcAttempts:20 grpcAttemptsDelay:2 singleActiveBackend:true parallelRequests:false} 11:41PM DBG Loading external backend: /build/backend/python/parler-tts/run.sh 11:41PM DBG Loading GRPC Process: /build/backend/python/parler-tts/run.sh 11:41PM DBG GRPC Service for will be running at: '127.0.0.1:36177' 11:41PM DBG GRPC Service state dir: /tmp/go-processmanager804239804 11:41PM DBG GRPC Service Started 11:41PM DBG GRPC(-127.0.0.1:36177): stdout Initializing libbackend for build 11:41PM DBG GRPC(-127.0.0.1:36177): stdout virtualenv activated 11:41PM DBG GRPC(-127.0.0.1:36177): stdout activated virtualenv has been ensured 11:41PM DBG [WatchDog] Watchdog checks for idle connections 11:41PM DBG GRPC(-127.0.0.1:36177): stderr /build/backend/python/parler-tts/venv/lib/python3.10/site-packages/transformers/utils/hub.py:124: FutureWarning: Using TRANSFORMERS_CACHE is deprecated and will be removed in v5 of Transformers. Use HF_HOME instead. 11:41PM DBG GRPC(-127.0.0.1:36177): stderr warnings.warn( 11:41PM DBG GRPC(-127.0.0.1:36177): stderr [parler-tts] startup: Namespace(addr='127.0.0.1:36177') 11:41PM DBG GRPC(-127.0.0.1:36177): stderr [parler-tts] Server started. Listening on: 127.0.0.1:36177 11:41PM DBG GRPC Service Ready 11:41PM DBG GRPC: Loading model with options: {state:{NoUnkeyedLiterals:{} DoNotCompare:[] DoNotCopy:[] atomicMessageInfo:} sizeCache:0 unknownFields:[] Model: ContextSize:512 Seed:2115209216 NBatch:512 F16Memory:false MLock:false MMap:true VocabOnly:false LowVRAM:false Embeddings:false NUMA:false NGPULayers:99999999 MainGPU: TensorSplit: Threads:6 LibrarySearchPath: RopeFreqBase:0 RopeFreqScale:0 RMSNormEps:0 NGQA:0 ModelFile:/models Device: UseTriton:false ModelBaseName: UseFastTokenizer:false PipelineType: SchedulerType: CUDA:false CFGScale:0 IMG2IMG:false CLIPModel: CLIPSubfolder: CLIPSkip:0 ControlNet: Tokenizer: LoraBase: LoraAdapter: LoraScale:0 NoMulMatQ:false DraftModel: AudioPath: Quantization: GPUMemoryUtilization:0 TrustRemoteCode:false EnforceEager:false SwapSpace:0 MaxModelLen:0 TensorParallelSize:0 MMProj: RopeScaling: YarnExtFactor:0 YarnAttnFactor:0 YarnBetaFast:0 YarnBetaSlow:0 Type: FlashAttention:false NoKVOffload:false} 11:41PM ERR Server error error="could not load model (no success): Unexpected err=OSError("Incorrect path_or_model_id: ''. Please provide either the path to a local folder or the repo_id of a model on the Hub."), type(err)=<class 'OSError'>" ip=10.0.1.224 latency=8.02050853s method=POST status=500 url=/tts

Additional context I tried running the latest AIO image without any of my usual volumes attached, environment variables etc. for a clean run and had the same result.