Spaces:
Runtime error
Runtime error
Commit
·
63b6d33
1
Parent(s):
4c6f82c
feat: Update Gradio interface configuration with description and transcript output
Browse files
README.md
CHANGED
|
@@ -55,77 +55,6 @@ To set up the project, follow these steps:
|
|
| 55 |
5. **Generate Audio:**
|
| 56 |
Click the button to start the conversion process. The output will be an MP3 file containing the podcast dialogue.
|
| 57 |
|
| 58 |
-
## Project Structure
|
| 59 |
-
|
| 60 |
-
- **main.py:** Main application script.
|
| 61 |
-
- **requirements.txt:** List of dependencies.
|
| 62 |
-
- **README.md:** Project documentation (this file).
|
| 63 |
-
|
| 64 |
-
## Code Explanation
|
| 65 |
-
|
| 66 |
-
### Dialogue Models
|
| 67 |
-
|
| 68 |
-
Defines the structure of the dialogue using Pydantic models.
|
| 69 |
-
|
| 70 |
-
```python
|
| 71 |
-
class DialogueItem(BaseModel):
|
| 72 |
-
text: str
|
| 73 |
-
voice: Literal["alloy", "onyx", "fable"]
|
| 74 |
-
|
| 75 |
-
class Dialogue(BaseModel):
|
| 76 |
-
scratchpad: str
|
| 77 |
-
dialogue: List[DialogueItem]
|
| 78 |
-
```
|
| 79 |
-
|
| 80 |
-
### LLM Function
|
| 81 |
-
|
| 82 |
-
Generates dialogue based on the input text using the `promptic` decorator.
|
| 83 |
-
|
| 84 |
-
```python
|
| 85 |
-
@llm(model="gemini/gemini-1.5-flash")
|
| 86 |
-
def generate_dialogue(text: str) -> Dialogue:
|
| 87 |
-
# Function to generate podcast dialogue
|
| 88 |
-
```
|
| 89 |
-
|
| 90 |
-
### TTS Function
|
| 91 |
-
|
| 92 |
-
Converts text to speech using OpenAI's text-to-speech model.
|
| 93 |
-
|
| 94 |
-
```python
|
| 95 |
-
def get_mp3(text: str, voice: str, api_key: str = None) -> bytes:
|
| 96 |
-
# Function to generate MP3 from text
|
| 97 |
-
```
|
| 98 |
-
|
| 99 |
-
### Main Function
|
| 100 |
-
|
| 101 |
-
Processes the PDF, generates dialogue, and converts it to audio.
|
| 102 |
-
|
| 103 |
-
```python
|
| 104 |
-
def generate_audio(file: bytes, openai_api_key: str) -> bytes:
|
| 105 |
-
# Main function to process PDF and generate audio
|
| 106 |
-
```
|
| 107 |
-
|
| 108 |
-
### Gradio Interface
|
| 109 |
-
|
| 110 |
-
Creates a user-friendly interface for uploading PDFs and generating podcasts.
|
| 111 |
-
|
| 112 |
-
```python
|
| 113 |
-
demo = gr.Interface(
|
| 114 |
-
title="PDF to Podcast",
|
| 115 |
-
description="Convert any PDF document into an engaging podcast episode!",
|
| 116 |
-
fn=generate_audio,
|
| 117 |
-
inputs=[
|
| 118 |
-
gr.File(label="Input PDF", type="binary"),
|
| 119 |
-
gr.Textbox(label="OpenAI API Key", placeholder="Enter your OpenAI API key here"),
|
| 120 |
-
],
|
| 121 |
-
outputs=[
|
| 122 |
-
gr.Audio(format="mp3"),
|
| 123 |
-
],
|
| 124 |
-
)
|
| 125 |
-
|
| 126 |
-
demo.launch(show_api=False)
|
| 127 |
-
```
|
| 128 |
-
|
| 129 |
## License
|
| 130 |
|
| 131 |
This project is licensed under the Apache 2.0 License. See the [LICENSE](LICENSE) file for more information.
|
|
|
|
| 55 |
5. **Generate Audio:**
|
| 56 |
Click the button to start the conversion process. The output will be an MP3 file containing the podcast dialogue.
|
| 57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
## License
|
| 59 |
|
| 60 |
This project is licensed under the Apache 2.0 License. See the [LICENSE](LICENSE) file for more information.
|