chore: Namespace serving child processes

lib/medical_transcription/application.ex

@@ -25,9 +25,9 @@ defmodule MedicalTranscription.Application do
       {Phoenix.PubSub, name: MedicalTranscription.PubSub},
       # Start the Finch HTTP client for sending emails
       {Finch, name: MedicalTranscription.Finch},
-      serving_child_spec(TranscriptionServing, transcription_serving, 4),
-      serving_child_spec(TokenClassificationServing, token_classification_serving, 1),
-      {Nx.Serving, serving: text_embedding_serving, name: TextEmbeddingServing},
+      serving_child_spec(MedicalTranscription.TranscriptionServing, transcription_serving, 4),
+      serving_child_spec(MedicalTranscription.TokenClassificationServing, token_classification_serving, 1),
+      {Nx.Serving, serving: text_embedding_serving, name: MedicalTranscription.TextEmbeddingServing},
       # Start a worker by calling: MedicalTranscription.Worker.start_link(arg)
       # {MedicalTranscription.Worker, arg},
       # Start to serve requests, typically the last entry

lib/medical_transcription/transcriber.ex

@@ -39,7 +39,7 @@ defmodule MedicalTranscription.Transcriber do
   end
 
   defp stream_transcription(audio_file_path) do
-    TranscriptionServing
+    MedicalTranscription.TranscriptionServing
     |> Nx.Serving.batched_run({:file, audio_file_path})
     |> Stream.with_index()
   end

lib/medical_transcription/utilities.ex

@@ -5,7 +5,7 @@ defmodule MedicalTranscription.Utilities do
 
   @doc "Creates a vector embedding for text using the text embedding serving in the application's supervision tree."
   def compute_vector_as_list(text) do
-    TextEmbeddingServing
+    MedicalTranscription.TextEmbeddingServing
     |> Nx.Serving.batched_run(text)
     |> Map.get(:embedding)
     |> Nx.to_flat_list()

lib/medical_transcription_web/components/transcription_text_component.ex

@@ -65,7 +65,9 @@ defmodule MedicalTranscriptionWeb.Components.TranscriptionTextComponent do
 
   defp find_keywords(live_view_pid, text) do
     # First, we use token classification to determine parts of speech and then retrieve the verb and adjective+noun phrases.
-    %{entities: entities} = Nx.Serving.batched_run(TokenClassificationServing, text)
+    %{entities: entities} =
+      Nx.Serving.batched_run(MedicalTranscription.TokenClassificationServing, text)
+
     phrases = AudioTagger.KeywordFinder.cleanup_phrases(entities)
 
     # Then, we use one of two processes to determine which to show as keywords: