app.py

import gradio as gr
import os
from dotenv import load_dotenv
from pydub import AudioSegment

load_dotenv()

from lang_list import TEXT_SOURCE_LANGUAGE_NAMES
from gradio_client import Client

HF_API = os.getenv("HF_API")
API_URL = os.getenv("API_URL")  # path to Seamlessm4t API endpoint

DEFAULT_TARGET_LANGUAGE = "Western Persian"

DESCRIPTION = """
# Seamlessm4t + Speaker Diarization + Voice Activity Detection
Here we use seamlessm4t to generate captions for full audios. Audio can be of arbitrary length. 
"""

DUPLICATE = """
To duplicate this repo, you have to give permission from three reopsitories and accept all user conditions: 

1- https://huggingface.co/pyannote/voice-activity-detection

2- https://hf.co/pyannote/segmentation

3- https://hf.co/pyannote/speaker-diarization

"""
from pyannote.audio import Pipeline
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization",use_auth_token=HF_API)
def predict(target_language, number_of_speakers, audio_source, input_audio_mic, input_audio_file):
    if audio_source == "microphone":
        input_data = input_audio_mic
    else:
        input_data = input_audio_file

    print(input_data)

    if number_of_speakers == 0:
        diarization = pipeline(input_data)
    else:
        diarization = pipeline(input_data, num_speakers=number_of_speakers)

    for turn, value, speaker in diarization.itertracks(yield_label=True):
        print(f"start={turn.start}s stop={turn.end}s speaker_{speaker}")

    song = AudioSegment.from_wav(input_data)

    client = Client(API_URL)
    output_text = ""
    for turn, value, speaker in diarization.itertracks(yield_label=True):
        print(turn)
        try:
            clipped = song[turn.start * 1000: turn.end * 1000]
            clipped.export(f'my.wav', format='wav', bitrate=16000)

            _, result = client.predict(
                "ASR (Automatic Speech Recognition)",
                "file",  # str in 'Audio source' Radio component
                f"my.wav",
                f"my.wav",
                "text",
                target_language,
                target_language,
                api_name="/run"
            )
            current_text = f'start: {turn.start:.1f} end: {turn.end:.1f} text: {result} speaker: {speaker}'
            yield current_text
            #if current_text is not None:
            #    output_text = output_text + "\n" + current_text


        except Exception as e:
            print(e)

        #return output_text

def update_audio_ui(audio_source: str) -> tuple[dict, dict]:
    mic = audio_source == "microphone"
    return (
        gr.update(visible=mic, value=None),  # input_audio_mic
        gr.update(visible=not mic, value=None),  # input_audio_file
    )


with gr.Blocks(css="style.css") as demo:
    gr.Markdown(DESCRIPTION)
    with gr.Group():
        with gr.Row():
            target_language = gr.Dropdown(
                choices=TEXT_SOURCE_LANGUAGE_NAMES,
                label="Output Language",
                value=DEFAULT_TARGET_LANGUAGE,
                interactive=True,
                info="Select your target language"
            )
            number_of_speakers=gr.Number(label="Number of Speakers",info="Keep it zero, if you want the model to automatically detect the number of speakers")
        with gr.Row() as audio_box:
            audio_source = gr.Radio(
                choices=["file", "microphone"], value="file", interactive=True
            )
            input_audio_mic = gr.Audio(
                label="Input speech",
                type="filepath",
                source="microphone",
                visible=False,
            )
            input_audio_file = gr.Audio(
                label="Input speech",
                type="filepath",
                source="upload",
                visible=True,
            )
            final_audio = gr.Audio(label="Output", visible=False)
        audio_source.change(
            fn=update_audio_ui,
            inputs=audio_source,
            outputs=[input_audio_mic, input_audio_file],
            queue=False,
            api_name=False,
        )
        input_audio_mic.change(lambda x: x, input_audio_mic, final_audio)
        input_audio_file.change(lambda x: x, input_audio_file, final_audio)
        submit = gr.Button("Submit")
        text_output = gr.Textbox(label="Transcribed Text", value="", interactive=False,lines=2,scale=3,max_lines=2)

        submit.click(fn=predict, inputs=[target_language,number_of_speakers, audio_source,input_audio_mic, input_audio_file], outputs=[text_output],api_name="predict")

    gr.Markdown(DUPLICATE)

demo.queue(max_size=50).launch()