Spaces:
Running
Running
File size: 2,428 Bytes
ee3575f bdf2bf2 c98669c bdf2bf2 c98669c bdf2bf2 ee3575f bdf2bf2 ee3575f c98669c ee3575f af4cf59 ee3575f c98669c ee3575f c98669c ee3575f |
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 |
import gradio as gr
import math
import time
import numpy as np
from pydub import AudioSegment
import io
def numpy_to_mp3(audio_array, sampling_rate):
# Normalize audio_array if it's floating-point
if np.issubdtype(audio_array.dtype, np.floating):
max_val = np.max(np.abs(audio_array))
audio_array = (audio_array / max_val) * 32767 # Normalize to 16-bit range
audio_array = audio_array.astype(np.int16)
# Create an audio segment from the numpy array
audio_segment = AudioSegment(
audio_array.tobytes(),
frame_rate=sampling_rate,
sample_width=audio_array.dtype.itemsize,
channels=1
)
# Export the audio segment to MP3 bytes - use a high bitrate to maximise quality
mp3_io = io.BytesIO()
audio_segment.export(mp3_io, format="mp3", bitrate="320k")
# Get the MP3 bytes
mp3_bytes = mp3_io.getvalue()
mp3_io.close()
return mp3_bytes
def stream(audio, chunk_length_s):
start_time = time.time()
sampling_rate, array = audio
chunk_length = int(chunk_length_s * sampling_rate)
time_length = chunk_length_s / 2 # always stream outputs faster than it takes to process
audio_length = len(array)
num_batches = math.ceil(audio_length / chunk_length)
for idx in range(num_batches):
time.sleep(time_length)
start_pos = idx * chunk_length
end_pos = min((idx + 1) * chunk_length, audio_length)
chunk = array[start_pos : end_pos]
chunk_mp3 = numpy_to_mp3(chunk, sampling_rate=sampling_rate)
if idx == 0:
first_time = round(time.time() - start_time, 2)
run_time = round(time.time() - start_time, 2)
yield chunk_mp3, first_time, run_time
with gr.Blocks() as demo:
with gr.Row():
with gr.Column():
audio_in = gr.Audio(value="librispeech.wav", sources=["upload"], type="numpy")
chunk_length = gr.Slider(minimum=2, maximum=10, value=2, step=2, label="Chunk length (s)")
run_button = gr.Button("Stream audio")
with gr.Column():
audio_out = gr.Audio(streaming=True, autoplay=True, format="mp3", label="mp3")
first_time = gr.Textbox(label="Time to first chunk (s)")
run_time = gr.Textbox(label="Time to current chunk (s)")
run_button.click(fn=stream, inputs=[audio_in, chunk_length], outputs=[audio_out, first_time, run_time])
demo.launch() |