Upload summarine.livemd

public-apps/summarine.livemd

@@ -0,0 +1,70 @@
+<!-- livebook:{"app_settings":{"show_source":true,"slug":"Summarine"}} -->
+
+# Summarine
+
+```elixir
+Mix.install(
+  [
+    {:kino_bumblebee, "~> 0.3.0"},
+    {:exla, "~> 0.5.1"},
+    {:req, "~> 0.3.11"}
+  ],
+  config: [nx: [default_backend: EXLA.Backend]]
+)
+```
+
+## Intro
+
+Audio to text, then summary.
+
+## App
+
+```elixir
+{:ok, model_info} = Bumblebee.load_model({:hf, "openai/whisper-base"})
+{:ok, featurizer} = Bumblebee.load_featurizer({:hf, "openai/whisper-base"})
+{:ok, tokenizer} = Bumblebee.load_tokenizer({:hf, "openai/whisper-base"})
+{:ok, generation_config} = Bumblebee.load_generation_config({:hf, "openai/whisper-base"})
+generation_config = Bumblebee.configure(generation_config, max_new_tokens: 100)
+
+serving =
+  Bumblebee.Audio.speech_to_text(model_info, featurizer, tokenizer, generation_config,
+    compile: [batch_size: 1],
+    defn_options: [compiler: EXLA]
+  )
+```
+
+```elixir
+audio_input = Kino.Input.audio("Audio", sampling_rate: featurizer.sampling_rate)
+form = Kino.Control.form([audio: audio_input], submit: "Run")
+frame = Kino.Frame.new()
+
+Kino.listen(form, fn %{data: %{audio: audio}} ->
+  if audio do
+    Kino.Frame.render(frame, Kino.Text.new("Running..."))
+
+    audio =
+      audio.data
+      |> Nx.from_binary(:f32)
+      |> Nx.reshape({:auto, audio.num_channels})
+      |> Nx.mean(axes: [1])
+
+    %{results: [%{text: generated_text}]} = Nx.Serving.run(serving, audio)
+    Kino.Frame.render(frame, Kino.Text.new(generated_text))
+  end
+end)
+
+Kino.Layout.grid([form, frame], boxed: true, gap: 16)
+```
+
+```elixir
+generated_text = "I'm blue and you are red, I'm very very blue."
+```
+
+```elixir
+payload = %{
+  model: "llama2-uncensored",
+  prompt: "Please summary the text: #{generated_text}"
+}
+
+Req.post!("http://localhost:11434/api/generate", json: payload).body
+```