Update README.md

This Hugging Face Space provides an interactive interface to use the Indic Parler-TTS model for generating high-quality, natural-sounding speech in Urdu, Punjabi, and Sindhi. Users can input text, customize speech attributes, and generate audio outputs with various styles, emotions, and speaker configurations.

Step 1: Select Language
Choose the language for your speech:

1: Urdu
2: Punjabi
3: Sindhi

Step 2: Configure Speaker Gender
Select the speaker's gender:

1: Male
2: Female

Step 3: Choose Emotion
Pick the emotion for the speech:

1: Neutral
2: Happy
3: Sad
4: Angry
5: Narration
6: Command
And more.

Step 4: Customize Speech Attributes
Fine-tune the speech by adjusting the following:

1: Background Noise: Clear or Slightly Noisy.
2: Reverberation: Close-Sounding or Distant-Sounding.
3: Expressivity: Expressive, Slightly Expressive, or Monotone.
4: Pitch: High, Low, or Balanced.
5: Speaking Rate: Slow, Moderate, or Fast.
6: Voice Quality: Basic or Refined.

Step 5: Enter the text you want to convert into speech in the text box.

Step 6: Generate and Edit the Description
Click the "Generate Caption/Description" button to generate a detailed description of the speech based on your configurations. The generated description appears in an editable textbox. You can edit this description to further refine the speech characteristics.

Step 7: Generate Speech
Once you're satisfied with the description, click the "Generate Speech" button. The Space will:

Use GPU resources to process the text and description.
Generate the corresponding audio file.
The audio file is displayed in the interface and can be played or downloaded.

README.md

@@ -11,4 +11,5 @@ license: apache-2.0
 short_description: IndicParler_TTS for Urdu_Punjabi & Sindhi
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Indic Parler-TTS: Multilingual Text-to-Speech (TTS) for Urdu, Punjabi, and Sindhi
+