File size: 2,626 Bytes
3b1f0f3
 
 
 
 
5bd7020
 
 
 
 
 
 
 
a7c380d
 
3b1f0f3
8a1ba2b
a368b88
8a1ba2b
770a448
8a1ba2b
8aa19f9
3b1f0f3
 
 
 
8aa19f9
3b1f0f3
cb3430d
3b1f0f3
 
 
a368b88
fc13aca
3b1f0f3
 
 
cb3430d
3b1f0f3
dce5f84
3b1f0f3
 
 
 
 
 
770a448
3b1f0f3
 
 
 
a7c380d
 
 
 
 
 
 
3b1f0f3
 
 
 
 
 
 
 
0ad3caf
 
3b1f0f3
 
a7c380d
3b1f0f3
 
 
 
 
 
 
 
 
dce5f84
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
import random 
import gradio as gr 
import numpy as np 
from elevenlabs import voices, generate, set_api_key, UnauthenticatedRateLimitError

def pad_buffer(audio):
    # Pad buffer to multiple of 2 bytes
    buffer_size = len(audio)
    element_size = np.dtype(np.int16).itemsize
    if buffer_size % element_size != 0:
        audio = audio + b'\0' * (element_size - (buffer_size % element_size))
    return audio 

def generate_voice(text, voice_name, api_key):
    set_api_key(api_key) #set API key
    try:
        audio = generate(
            text[:4000], # Limit to 4000 characters
            voice=voice_name, 
            model="eleven_multilingual_v2"
        )
        return (44100, np.frombuffer(pad_buffer(audio), dtype=np.int16))
    except UnauthenticatedRateLimitError as e:
        raise gr.Error("Thanks for trying out ElevenLabs TTS! You've reached the free tier limit. Please provide an API key to continue.") 
    except Exception as e:
        raise gr.Error(e)
    
description = """
Eleven Multilingual V2 is the world's best Text-to-Speech model. Sign up on [ElevenLabs](elevenlabs.io/?from=partnerpierce7156) to get an API Key.
"""

with gr.Blocks() as block:
    #gr.Markdown('[ ![ElevenLabs](https://user-images.githubusercontent.com/12028621/262629275-4f85c9cf-85b6-435e-ab50-5b8c7c4e9dd2.png) ](https://elevenlabs.io)')
    gr.Markdown("# <center> ElevenLabs TTS </center>")
    gr.Markdown(description)
    
    input_text = gr.Textbox(
        label="Input Text (4000 characters max)", 
        lines=2, 
        value="Hello! 你好! Hola! नमस्ते! Bonjour! こんにちは! مرحبا! 안녕하세요! Ciao! Cześć! Привіт! Γειά σας! Здравей! வணக்கம்!",
        elem_id="input_text"
    )

    all_voices = voices() 
    input_voice = gr.Dropdown(
        [ voice.name for voice in all_voices ], 
        value="Bella",
        label="Voice", 
        elem_id="input_voice"
    )

    input_api_key = gr.Textbox(
    type='password', 
    label='Elevenlabs API Key', 
    placeholder='Enter your API key',
    elem_id="input_api_key"
    )
    
    run_button = gr.Button(
        text="Generate Voice", 
        type="button"
    )

    out_audio = gr.Audio(
        label="Generated Voice",
        type="numpy", 
        elem_id="out_audio",
        format="mp3"
    )
        
    inputs = [input_text, input_voice, input_api_key]
    outputs = [out_audio]
    
    run_button.click(
        fn=generate_voice, 
        inputs=inputs, 
        outputs=outputs, 
        queue=True
    )

block.queue(concurrency_count=5).launch(debug=True)