File size: 1,844 Bytes
5f06acc 03f2172 5f06acc 03f2172 5f06acc 03f2172 5f06acc 03f2172 5f06acc 497d091 5f06acc 03f2172 5f06acc 497d091 |
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 |
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):
model_name = "eleven_multilingual_v1"
try:
audio = generate(
text[:250], # Limit to 250 characters
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))
with gr.Blocks() as block:
input_text = gr.Textbox(
label="Input Text (250 characters max)",
lines=2,
value="Hahaha OHH MY GOD! This is SOOO funny, I-I am Eleven a text-to-speech system!",
elem_id="input_text"
)
all_voices = voices()
input_voice = gr.Dropdown(
[voice.name for voice in all_voices],
value="Arnold",
label="Voice",
elem_id="input_voice"
)
run_button = gr.Button()
out_audio = gr.Audio(
label="Generated Voice",
type="numpy",
elem_id="out_audio"
)
inputs = [input_text, input_voice]
outputs = [out_audio]
run_button.click(
fn=generate_voice,
inputs=inputs,
outputs=outputs,
queue=True
)
block.queue(concurrency_count=1).launch(debug=True, theme="Monochrome") |