Spaces:
Running
Running
File size: 5,311 Bytes
42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed 42966de dfb71ed |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 |
"""
This script implements a Gradio interface for text-to-speech conversion using OpenAI's API.
Users can input text, select a model and voice, and receive an audio output of the synthesized speech.
Dependencies:
- gradio
- openai
Usage:
Run the script to launch a web interface for text-to-speech conversion.
Note:
- Ensure that you have installed the required packages:
pip install gradio openai
- Obtain a valid OpenAI API key with access to the necessary services.
"""
import gradio as gr
import tempfile
import openai
from typing import List
def tts(input_text: str, model: str, voice: str, api_key: str) -> str:
"""
Convert input text to speech using OpenAI's Text-to-Speech API.
:param input_text: The text to be converted to speech.
:type input_text: str
:param model: The model to use for synthesis (e.g., 'tts-1', 'tts-1-hd').
:type model: str
:param voice: The voice profile to use (e.g., 'alloy', 'echo', 'fable', etc.).
:type voice: str
:param api_key: OpenAI API key.
:type api_key: str
:return: File path to the generated audio file.
:rtype: str
:raises ValueError: If input parameters are invalid.
:raises openai.error.OpenAIError: If API call fails.
"""
if not input_text.strip():
raise ValueError("Input text cannot be empty.")
if not api_key.strip():
raise ValueError("API key is required.")
openai.api_key = api_key
try:
response = openai.Audio.create(
text=input_text,
voice=voice,
model=model
)
except openai.error.OpenAIError as e:
raise e
if not hasattr(response, 'audio'):
raise Exception("Invalid response from OpenAI API. The response does not contain audio content.")
with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
temp_file.write(response.audio)
temp_file_path = temp_file.name
return temp_file_path
def tts_batch(
input_texts: List[str],
model: str,
voice: str,
api_key: str,
) -> List[str]:
"""
Convert a batch of input texts to speech using OpenAI's Text-to-Speech API.
:param input_texts: The texts to be converted to speech.
:type input_texts: List[str]
:param model: The model to use for synthesis.
:type model: str
:param voice: The voice profile to use.
:type voice: str
:param api_key: OpenAI API key.
:type api_key: str
:return: List of file paths to the generated audio files.
:rtype: List[str]
"""
outputs = []
for input_text in input_texts:
try:
output = tts(input_text, model, voice, api_key)
outputs.append(output)
except Exception as e:
outputs.append(None)
return outputs
def main():
"""
Main function to create and launch the Gradio interface with API capabilities and enhancements.
"""
# Define model and voice options
MODEL_OPTIONS = ["tts-1", "tts-1-hd"]
VOICE_OPTIONS = ["alloy", "echo", "fable", "onyx", "nova", "shimmer"]
with gr.Blocks() as demo:
with gr.Row():
with gr.Column(scale=1):
api_key_input = gr.Textbox(
label="API Key", type="password", placeholder="Enter your OpenAI API Key"
)
model_dropdown = gr.Dropdown(
choices=MODEL_OPTIONS, label="Model", value="tts-1"
)
voice_dropdown = gr.Dropdown(
choices=VOICE_OPTIONS, label="Voice Options", value="echo"
)
with gr.Column(scale=2):
input_textbox = gr.Textbox(
label="Input Text",
lines=10,
placeholder="Type your text here..."
)
submit_button = gr.Button("Convert Text to Speech", variant="primary")
with gr.Column(scale=1):
output_audio = gr.Audio(label="Output Audio")
error_output = gr.Textbox(
label="Error Message", interactive=False, visible=False
)
# Define the event handler for the submit button with API endpoint and naming
submit_button.click(
fn=tts,
inputs=[input_textbox, model_dropdown, voice_dropdown, api_key_input],
outputs=output_audio,
api_name="convert_text_to_speech",
)
# Allow pressing Enter in the input textbox to trigger the conversion
input_textbox.submit(
fn=tts,
inputs=[input_textbox, model_dropdown, voice_dropdown, api_key_input],
outputs=output_audio,
api_name="convert_text_to_speech",
)
# Expose the `demo` app as a callable function
def process_text_to_speech(
input_text: str,
model: str = "tts-1",
voice: str = "echo",
api_key: str = ""
) -> str:
"""
Allows calling the Gradio app as a function.
"""
return demo(
input_text,
model,
voice,
api_key
)
# Launch the Gradio app with API documentation enabled
demo.launch(share=True)
if __name__ == "__main__":
main() |