Update README.md

README.md

@@ -9,5 +9,20 @@ app_file: app.py
 pinned: false
 license: mit
 ---
+1. Libraries and Tools Used:
+    - Transformers: Provides the `VitsModel` and `AutoTokenizer`, with the use of `facebook/mms-tts-eng` model, a sophisticated text-to-speech model designed by Facebook.
+    - Torch: A companion library for Transformers, essential for processing the data through the speech model.
+    - Librosa: A library for audio processing, especially used here for pitch adjustment of the speech.
+    - Soundfile: Utilized to save the speech output as an audio file.
+    - Tempfile: Creates temporary files for intermediate storage during processing.
+    - Gradio: Facilitates the creation of a user-friendly web interface for the text-to-speech application.
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+2. Pipeline for Text-to-Speech Conversion:
+    - Text Input: You begin by typing in the text you want to be converted into speech.
+    - Tokenization: `AutoTokenizer` processes this text, preparing it for the speech model.
+    - Speech Synthesis: The `facebook/mms-tts-eng` model within the `VitsModel` takes this processed text and generates the spoken words.
+    - Pitch Adjustment: 0 Pitch Value: Represents the normal, unaltered pitch of the speech. This is the default state where the voice sounds as it naturally would, without any modifications.
+     Negative Pitch Values: When you set the pitch to a negative value, it makes the voice sound higher. This is similar to moving up the notes on a piano, resulting in a higher, perhaps more youthful or feminine tone.
+     Positive Pitch Values: Conversely, positive pitch values make the voice sound lower. This is akin to moving down the notes on a piano. A positive pitch shift results in a deeper, more resonant tone, often associated with a more masculine or mature voice.
+    - Saving Audio: The speech with the adjusted pitch is saved as an audio file using `Soundfile` and `Tempfile`.
+    - Interactive Web Interface: Gradio provides an interface where you input text, adjust the pitch using a slider, and listen to the speech output.