Edge TTS: The Ultimate Guide for Developers

1pip install edge-tts
2

1pip install --upgrade pip
2

1import asyncio
2import edge_tts
3
4async def main():
5    text = "Hello, Edge TTS! Installation successful."
6    voice = "en-US-JennyNeural"
7    output_file = "output.mp3"
8
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        print(f"Audio saved to {output_file}")
13    except Exception as e:
14        print(f"Error: {e}")
15
16if __name__ == "__main__":
17    asyncio.run(main())
18

1python test_edge_tts.py
2

1import asyncio
2import edge_tts
3
4async def main():
5    text = "This is a basic example of Edge TTS."
6    voice = "en-US-JennyNeural"  # Or any other available voice
7    output_file = "basic_output.mp3"
8
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        print(f"Audio saved to {output_file}")
13    except Exception as e:
14        print(f"Error: {e}")
15
16if __name__ == "__main__":
17    asyncio.run(main())
18

1import asyncio
2import edge_tts
3
4async def list_voices():
5    voices = await edge_tts.VoicesManager.create()
6    for voice in voices.find(language="en"): # You can filter based on language
7        print(voice)
8
9if __name__ == "__main__":
10    asyncio.run(list_voices())
11

1import asyncio
2import edge_tts
3
4async def main():
5    text = "Hello, I am using a different voice!"
6    voice = "en-GB-MiaNeural" # British English
7    output_file = "voice_output.mp3"
8
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        print(f"Audio saved to {output_file}")
13    except Exception as e:
14        print(f"Error: {e}")
15
16if __name__ == "__main__":
17    asyncio.run(main())
18

1import asyncio
2import edge_tts
3
4async def main():
5    text = '<speak><prosody rate="x-slow" volume="x-loud" pitch="+20Hz">This is a slow, loud, and high-pitched voice.</prosody> <expressive style="cheerful">And this is a cheerful voice!</expressive></speak>'
6    voice = "en-US-JennyNeural"
7    output_file = "advanced_output.mp3"
8
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        print(f"Audio saved to {output_file}")
13    except Exception as e:
14        print(f"Error: {e}")
15
16if __name__ == "__main__":
17    asyncio.run(main())
18

1import asyncio
2import edge_tts
3
4async def main():
5    text = "This text might cause an error."
6    voice = "invalid-voice" # intentionally invalid voice
7    output_file = "error_output.mp3"
8
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        print(f"Audio saved to {output_file}")
13    except Exception as e:
14        print(f"Error occurred: {type(e).__name__} - {e}")
15
16if __name__ == "__main__":
17    asyncio.run(main())
18

1import gradio as gr
2import asyncio
3import edge_tts
4import os
5
6async def generate_speech(text, voice):
7    output_file = "output.mp3"
8    try:
9        communicate = edge_tts.Communicate(text, voice)
10        await communicate.save(output_file)
11        return output_file # Return the audio file path
12    except Exception as e:
13        return str(e)
14
15async def list_voices():
16    voices = await edge_tts.VoicesManager.create()
17    voice_options = [voice["Name"] for voice in voices.data]
18    return voice_options
19
20async def main():
21  voice_options = await list_voices()
22  if isinstance(voice_options, str):
23    iface = gr.Interface(
24        fn=lambda x: voice_options,
25        inputs="text",
26        outputs="text",
27        title="Error Retrieving Voices",
28    )
29  else:
30      iface = gr.Interface(
31          fn=generate_speech,
32          inputs=["text", gr.Dropdown(voice_options, label="Voice")],
33          outputs="audio",
34          title="Edge TTS Web UI",
35          description="Enter text and select a voice to generate speech.",
36      )
37
38  iface.launch()
39
40if __name__ == "__main__":
41    asyncio.run(main())
42

1import tkinter as tk
2from tkinter import ttk
3import asyncio
4import edge_tts
5import threading
6
7async def speak(text, voice):
8    output_file = "output.mp3"
9    try:
10        communicate = edge_tts.Communicate(text, voice)
11        await communicate.save(output_file)
12        # Play audio - platform dependent, using a simple approach for example
13        os.system(f"start {output_file}") # For Windows; use appropriate command for other OS
14    except Exception as e:
15        print(f"Error: {e}")
16
17def on_button_click():
18    text = text_entry.get()
19    voice = voice_combobox.get()
20    threading.Thread(target=lambda: asyncio.run(speak(text, voice))).start()
21
22async def get_voices():
23    voices = await edge_tts.VoicesManager.create()
24    return [voice["Name"] for voice in voices.data]
25
26async def main():
27    global text_entry, voice_combobox
28    root = tk.Tk()
29    root.title("Edge TTS GUI")
30
31    text_label = ttk.Label(root, text="Enter Text:")
32    text_label.pack()
33
34    text_entry = ttk.Entry(root, width=50)
35    text_entry.pack()
36
37    voice_label = ttk.Label(root, text="Select Voice:")
38    voice_label.pack()
39
40    voice_list = await get_voices()
41    voice_combobox = ttk.Combobox(root, values=voice_list, state="readonly")
42    voice_combobox.pack()
43    voice_combobox.set(voice_list[0]) # Default voice
44
45    speak_button = ttk.Button(root, text="Speak", command=on_button_click)
46    speak_button.pack()
47
48    root.mainloop()
49
50if __name__ == "__main__":
51    os.environ['PY_IGNORE_ERROR']= "1" #ignore error caused by os.start
52    asyncio.run(main())
53