new entry

car_assistant_slim.ipynb

@@ -19,13 +19,13 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/opt/conda/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "/opt/homebrew/Caskroom/miniconda/base/envs/llm/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
       "  from .autonotebook import tqdm as notebook_tqdm\n",
-      "/opt/conda/lib/python3.10/site-packages/transformers/utils/generic.py:441: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
+      "/opt/homebrew/Caskroom/miniconda/base/envs/llm/lib/python3.11/site-packages/transformers/utils/generic.py:441: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
       "  _torch_pytree._register_pytree_node(\n",
-      "/opt/conda/lib/python3.10/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
+      "/opt/homebrew/Caskroom/miniconda/base/envs/llm/lib/python3.11/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
       "  _torch_pytree._register_pytree_node(\n",
-      "/opt/conda/lib/python3.10/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
+      "/opt/homebrew/Caskroom/miniconda/base/envs/llm/lib/python3.11/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
       "  _torch_pytree._register_pytree_node(\n"
      ]
     }
@@ -116,21 +116,35 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
+      "preprocessor_config.json: 100%|██████████| 185k/185k [00:00<00:00, 94.3MB/s]\n",
+      "tokenizer_config.json: 100%|██████████| 283k/283k [00:00<00:00, 1.05MB/s]\n",
+      "vocab.json: 100%|██████████| 836k/836k [00:00<00:00, 3.03MB/s]\n",
+      "tokenizer.json: 100%|██████████| 2.48M/2.48M [00:00<00:00, 50.6MB/s]\n",
+      "merges.txt: 100%|██████████| 494k/494k [00:00<00:00, 28.8MB/s]\n",
+      "normalizer.json: 100%|██████████| 52.7k/52.7k [00:00<00:00, 67.8MB/s]\n",
+      "added_tokens.json: 100%|██████████| 34.6k/34.6k [00:00<00:00, 38.7MB/s]\n",
+      "special_tokens_map.json: 100%|██████████| 2.19k/2.19k [00:00<00:00, 8.88MB/s]\n",
+      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
+      "config.json: 100%|██████████| 1.97k/1.97k [00:00<00:00, 4.46MB/s]\n",
+      "model.safetensors: 100%|██████████| 967M/967M [00:12<00:00, 74.9MB/s] \n",
+      "generation_config.json: 100%|██████████| 3.87k/3.87k [00:00<00:00, 39.0MB/s]\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " > tts_models/multilingual/multi-dataset/xtts_v1.1 is already downloaded.\n"
+      " > Downloading model to /Users/sasan.jafarnejad/Library/Application Support/tts/tts_models--multilingual--multi-dataset--xtts_v1.1\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/opt/conda/lib/python3.10/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
+      "100%|██████████| 1.87G/1.87G [00:24<00:00, 75.6MiB/s]\n",
+      "100%|██████████| 4.70k/4.70k [00:00<00:00, 17.9kiB/s]\n",
+      "100%|██████████| 294k/294k [00:00<00:00, 1.23MiB/s]\n",
+      "/opt/homebrew/Caskroom/miniconda/base/envs/llm/lib/python3.11/site-packages/transformers/utils/generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.\n",
       "  _torch_pytree._register_pytree_node(\n"
      ]
     },
@@ -138,6 +152,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      " > Model's license - CPML\n",
+      " > Check https://coqui.ai/cpml.txt for more info.\n",
       " > Using model: xtts\n"
      ]
     }
@@ -160,7 +176,20 @@
     "collapsed": true,
     "id": "JNALTDb0LT90"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "config.json: 100%|██████████| 1.42k/1.42k [00:00<00:00, 3.24MB/s]\n",
+      "model.safetensors: 100%|██████████| 1.11G/1.11G [00:13<00:00, 79.5MB/s]\n",
+      "tokenizer_config.json: 100%|██████████| 502/502 [00:00<00:00, 5.01MB/s]\n",
+      "sentencepiece.bpe.model: 100%|██████████| 5.07M/5.07M [00:00<00:00, 78.4MB/s]\n",
+      "tokenizer.json: 100%|██████████| 9.08M/9.08M [00:00<00:00, 61.5MB/s]\n",
+      "special_tokens_map.json: 100%|██████████| 239/239 [00:00<00:00, 372kB/s]\n"
+     ]
+    }
+   ],
    "source": [
     "#load model language recognition\n",
     "model_ckpt = \"papluca/xlm-roberta-base-language-detection\"\n",
@@ -183,8 +212,10 @@
    "source": [
     "#load model llama2\n",
     "mn = 'stabilityai/StableBeluga-7B' #mn = \"TheBloke/Llama-2-7b-Chat-GPTQ\" --> other possibility \n",
-    "model = AutoModelForCausalLM.from_pretrained(mn, device_map=0, load_in_4bit=True) #torch_dtype=torch.float16\n",
-    "tokr = AutoTokenizer.from_pretrained(mn, load_in_4bit=True) #tokenizer"
+    "# model = AutoModelForCausalLM.from_pretrained(mn, device_map=0, load_in_4bit=True) #torch_dtype=torch.float16\n",
+    "model = AutoModelForCausalLM.from_pretrained(mn, device_map=0) #torch_dtype=torch.float16\n",
+    "# tokr = AutoTokenizer.from_pretrained(mn, load_in_4bit=True) #tokenizer\n",
+    "tokr = AutoTokenizer.from_pretrained(mn) #tokenizer"
    ]
   },
   {
@@ -881,7 +912,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.13"
+   "version": "3.11.8"
   }
  },
  "nbformat": 4,

kitt.py

@@ -0,0 +1,195 @@
+import gradio as gr
+import plotly.express as px
+import requests
+
+# INTERFACE WITH AUDIO TO AUDIO
+
+
+def calculate_route():
+    api_key = "api_key"
+    origin = "49.631997,6.171029"
+    destination = "49.586745,6.140002"
+
+    url = f"https://api.tomtom.com/routing/1/calculateRoute/{origin}:{destination}/json?key={api_key}"
+    response = requests.get(url)
+    data = response.json()
+
+    lats = []
+    lons = []
+
+    for point in data['routes'][0]['legs'][0]['points']:
+        lats.append(point['latitude'])
+        lons.append(point['longitude'])
+    # fig = px.line_geo(lat=lats, lon=lons)
+    # fig.update_geos(fitbounds="locations")
+
+    fig = px.line_mapbox(lat=lats, lon=lons, zoom=12, height=600)
+
+    fig.update_layout(mapbox_style="open-street-map", mapbox_zoom=12, mapbox_center_lat=lats[0], mapbox_center_lon=lons[0])
+    fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
+
+    return fig
+
+
+def transcript(
+    general_context, link_to_audio, voice, emotion, place, time, delete_history, state
+):
+    """this function manages speech-to-text to input Fnanswer function and text-to-speech with the Fnanswer output"""
+    # load audio from a specific path
+    audio_path = link_to_audio
+    audio_array, sampling_rate = librosa.load(
+        link_to_audio, sr=16000
+    )  # "sr=16000" ensures that the sampling rate is as required
+
+    # process the audio array
+    input_features = processor(
+        audio_array, sampling_rate, return_tensors="pt"
+    ).input_features
+    predicted_ids = modelw.generate(input_features)
+    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+    quest_processing = FnAnswer(
+        general_context, transcription, place, time, delete_history, state
+    )
+    state = quest_processing[2]
+    print("langue " + quest_processing[3])
+
+    tts.tts_to_file(
+        text=str(quest_processing[0]),
+        file_path="output.wav",
+        speaker_wav=f"Audio_Files/{voice}.wav",
+        language=quest_processing[3],
+        emotion="angry",
+    )
+
+    audio_path = "output.wav"
+    return audio_path, state["context"], state
+
+
+# to be able to use the microphone on chrome, you will have to go to chrome://flags/#unsafely-treat-insecure-origin-as-secure and enter http://10.186.115.21:7860/
+# in "Insecure origins treated as secure", enable it and relaunch chrome
+
+# example question:
+# what's the weather like outside?
+# What's the closest restaurant from here?
+
+
+import gradio as gr
+
+shortcut_js = """
+<script>
+function shortcuts(e) {
+    var event = document.all ? window.event : e;
+    switch (e.target.tagName.toLowerCase()) {
+        case "input":
+        case "textarea":
+        break;
+        default:
+        if (e.key.toLowerCase() == "r" && e.ctrlKey) {
+            console.log("recording")
+            document.getElementById("recorder").start_recording();
+        }
+        if (e.key.toLowerCase() == "s" && e.ctrlKey) {
+            console.log("stopping")
+            document.getElementById("recorder").stop_recording();
+        }
+    }
+}
+document.addEventListener('keypress', shortcuts, false);
+</script>
+"""
+
+# with gr.Blocks(head=shortcut_js) as demo:
+#     action_button = gr.Button(value="Name", elem_id="recorder")
+#     textbox = gr.Textbox()
+#     action_button.click(lambda : "button pressed", None, textbox)
+
+# demo.launch()
+
+
+# Generate options for hours (00-23)
+hour_options = [f"{i:02d}:00:00" for i in range(24)]
+
+model_answer = ""
+general_context = ""
+# Define the initial state with some initial context.
+print(general_context)
+initial_state = {"context": general_context}
+initial_context = initial_state["context"]
+# Create the Gradio interface.
+
+
+with gr.Blocks(theme=gr.themes.Default()) as demo:
+
+    with gr.Row():
+        with gr.Column(scale=1, min_width=300):
+            time_picker = gr.Dropdown(
+                choices=hour_options, label="What time is it?", value="08:00:00"
+            )
+            history = gr.Radio(
+                ["Yes", "No"], label="Maintain the conversation history?", value="No"
+            )
+            voice_character = gr.Radio(
+                choices=[
+                    "Rick Sanches",
+                    "Eddie Murphy",
+                    "David Attenborough",
+                    "Morgan Freeman",
+                ],
+                label="Choose a voice",
+                value="Rick Sancher",
+                show_label=True,
+            )
+            emotion = gr.Radio(
+                choices=["Cheerful", "Grumpy"],
+                label="Choose an emotion",
+                value="Cheerful",
+                show_label=True,
+            )
+            # place = gr.Radio(
+            #     choices=[
+            #         "Luxembourg Gare, Luxembourg",
+            #         "Kirchberg Campus, Kirchberg",
+            #         "Belval Campus, Belval",
+            #         "Eiffel Tower, Paris",
+            #         "Thionville, France",
+            #     ],
+            #     label="Choose a location for your car",
+            #     value="Kirchberg Campus, Kirchberg",
+            #     show_label=True,
+            # )
+            origin = gr.Textbox(value="Luxembourg Gare, Luxembourg", label="Origin", interactive=True)
+            destination = gr.Textbox(
+                value="Kirchberg Campus, Kirchberg", label="Destination", interactive=True)
+            recorder = gr.Audio(type="filepath", label="input audio", elem_id="recorder")
+        with gr.Column(scale=2, min_width=600):
+            map_plot = gr.Plot()
+            origin.submit(fn=calculate_route, outputs=map_plot)
+            destination.submit(fn=calculate_route, outputs=map_plot)
+            output_audio = gr.Audio(label="output audio")
+            # map_if = gr.Interface(fn=plot_map, inputs=year_input, outputs=map_plot)
+
+    # iface = gr.Interface(
+    #     fn=transcript,
+    #     inputs=[
+    #         gr.Textbox(value=initial_context, visible=False),
+    #         gr.Audio(type="filepath", label="input audio", elem_id="recorder"),
+    #         voice_character,
+    #         emotion,
+    #         place,
+    #         time_picker,
+    #         history,
+    #         gr.State(),  # This will keep track of the context state across interactions.
+    #     ],
+    #     outputs=[gr.Audio(label="output audio"), gr.Textbox(visible=False), gr.State()],
+    #     head=shortcut_js,
+    # )
+
+# close all interfaces open to make the port available
+gr.close_all()
+# Launch the interface.
+
+demo.queue().launch(
+    debug=True, server_name="0.0.0.0", server_port=7860, ssl_verify=False
+)
+
+# iface.launch(debug=True, share=False, server_name="0.0.0.0", server_port=7860, ssl_verify=False)