File size: 1,389 Bytes
5f06acc 5a75523 03f2172 5f06acc 00bbc0d 5f06acc 03f2172 5f06acc 5a75523 dda0cbf 5a75523 f459688 00bbc0d 562ed19 5a75523 562ed19 1e717a6 f459688 562ed19 5f06acc 9a0de0e |
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 |
import gradio as gr
import numpy as np
from elevenlabs import voices, generate, set_api_key, UnauthenticatedRateLimitError
def pad_buffer(audio):
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):
model_name = "eleven_multilingual_v1"
try:
audio = generate(
text[:250],
voice=voice_name,
model=model_name
)
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(str(e))
all_voices = voices()
desired_voices = ["Antonio"]
filtered_voices = [voice.name for voice in all_voices if voice.name in desired_voices]
input_text = gr.Textbox(label="Input Text", lines=2)
input_voice = gr.Dropdown(choices=filtered_voices, default="Antonio", label="Voice")
out_audio = gr.Audio(label="Generated Voice", type="numpy")
iface = gr.Interface(
fn=generate_voice,
inputs=[input_text, input_voice],
outputs=out_audio,
theme="Monochrome",
live=True
)
iface.launch() |