adding the docs

app.py

@@ -1,11 +1,35 @@
 import gradio as gr
-import librosa
 import openai
 from constants import *
 
 
 openai.api_key = OPENAI_API_KEY
 
+title = "Car Seats Voice Commands"
+
+description = """
+This is a demo for controlling car seats with Voice Commands, On the left there's the inputs section
+and on the right you'll find your outputs. For the inputs you have two choices **Voice** and **Text**,
+Use **Voice** If you want a closer experience to the final product, Or use **Text** if you just want to test the command model.
+for the outputs you have the **transcription**(Please check that it's accurate), **command**(to know which
+command the system detected) and you have the robot voice (again use this if you want a more real experience).
+
+**Features** : You can either activate of deactivate the following features 
+- Heated Seats 
+- Cooled Seats 
+- Massage Seats 
+
+Examples:
+- **Direct Commands** : Try to say something like "Activate heated seats" or "Turn Off massage seats"
+- **Indirect Commands** : Try "My back is cold" , "No heating is needed anymore" or "I'm stressed today"
+"""
+
+article = """
+This demo processes commands in two steps, the first step is the transcription phase and the second is the
+Command Classification phase. For Transcription I used The OpenAi whisper model, and for the classification 
+I Fine-Tuned the OpenAi **ada** model on Car Seats Command.
+"""
+
 
 def get_command(command, model, id2label):
     """
@@ -33,16 +57,6 @@ def transcribe(audio, text):
     audio_file = open(audio, "rb")
     transcription = openai.Audio.transcribe("whisper-1", audio_file, language="en")
     transcription = transcription["text"]
-    # Downsample original frequency to 16000hrz
-    # input, rate = librosa.load(audio, sr=16000)
-
-    # getting text transcription
-    # inputs = processor(input, sampling_rate=rate, return_tensors="pt")
-    # generated_ids = model.generate(
-    #     inputs["input_features"], attention_mask=inputs["attention_mask"]
-    # )
-
-    # transcription = processor.batch_decode(generated_ids, skip_special_tokens=True)
 
     result = get_command(transcription, MODEL, id2label)
     audio_res = resoponses.get(result)()
@@ -65,4 +79,7 @@ if __name__ == "__main__":
             gr.Audio(label="Machine Response (Audio Version)"),
         ],
         allow_flagging="auto",
+        title=title,
+        description=description,
+        article=article,
     ).launch()

constants.py

@@ -1,6 +1,5 @@
 from commands import ResponseManager
 from dotenv import load_dotenv
-from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration
 import os
 
 load_dotenv()
@@ -36,9 +35,3 @@ text_respnses = {
     "cooled_seats_on": "Cooled seats activated",
     "cooled_seats_off": "Cooled seats deactivated",
 }
-
-# model = Speech2TextForConditionalGeneration.from_pretrained(
-#     "facebook/s2t-small-librispeech-asr"
-# )
-
-# processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr")

flagged/log.csv

@@ -31,3 +31,8 @@ C:\Users\admin\Desktop\DS JRP\Programming\Python\Portfolio\Voice_Commands\flagge
 /home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmp68h39ev5.wav,,,,2023-03-03 11:54:26.488504
 ,I'm cold,,,2023-03-04 10:42:40.559129
 ,I'm cold,,,2023-03-04 10:44:49.418487
+/home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmpz0hq94rg.wav,,,,2023-03-04 11:32:53.276143
+/home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmpr_cd3uqa.wav,,,,2023-03-04 11:33:22.605201
+/home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmp8pxcjozd.wav,,,,2023-03-04 11:34:02.384882
+/home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmp7vog58pt.wav,,,,2023-03-04 11:34:32.302966
+/home/boumaza/Desktop/Programming/Python/Voice_Commands/flagged/component 0/tmp5v5ipo64.wav,,,,2023-03-04 11:34:50.524003