app.py

from __future__ import annotations

import os

import gradio as gr
import numpy as np
import torch
import torchaudio
from seamless_communication.models.inference import Translator

from lang_list import LANGUAGE_NAME_TO_CODE, S2TT_TARGET_LANGUAGE_NAMES 


DESCRIPTION = """# Speech to Text Translation 

[SeamlessM4T](https://github.com/facebookresearch/seamless_communication) is designed to provide high-quality
translation, allowing people from different linguistic communities to communicate effortlessly through speech and text.

This unified model enables multiple tasks like Speech-to-Speech (S2ST), Speech-to-Text (S2TT), Text-to-Speech (T2ST)
translation and more, without relying on multiple separate models. Here the task is to do the speech to text translation. 
"""

CACHE_EXAMPLES = os.getenv("CACHE_EXAMPLES") == "1"


AUDIO_SAMPLE_RATE  = 44100
#MAX_INPUT_AUDIO_LENGTH =  1800 # in seconds
DEFAULT_TARGET_LANGUAGE = "French"

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
translator = Translator(
    model_name_or_card="seamlessM4T_medium",
    vocoder_name_or_card="vocoder_36langs",
    device=device,
)

task_name = "S2TT (Speech to Text translation)"
task_name = task_name.split()[0]

def predict(
    audio_source: str,
    input_audio_mic: str | None,
    input_audio_file: str | None,
    target_language: str,
) -> str:
    target_language_code = LANGUAGE_NAME_TO_CODE[target_language]

    if audio_source == "microphone": 
        input_data = input_audio_mic
    else:
        input_data = input_audio_file

    arr, org_sr = torchaudio.load(input_data)
    new_arr = torchaudio.functional.resample(arr, orig_freq=org_sr, new_freq=AUDIO_SAMPLE_RATE)
   
    torchaudio.save(input_data, new_arr, sample_rate=int(AUDIO_SAMPLE_RATE))

    text_out, wav, sr = translator.predict(
        input = input_data,
        task_str = task_name,
        tgt_lang=target_language_code,
        ngram_filtering=True,
    )
    
    return text_out



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(
                label="Target language",
                choices=S2TT_TARGET_LANGUAGE_NAMES,
            )
            
        with gr.Row() as audio_box:
        
            audio_source = gr.Radio(
                label="Audio source",
                choices=["file", "microphone"],
                value="file",
            )
            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,
            )
                
        btn = gr.Button("Translate") 
        with gr.Column():
            output_text = gr.Textbox(label="Translated text")

        audio_source.change(
            fn=update_audio_ui,
            inputs=audio_source,
            outputs=[
                input_audio_mic,
                input_audio_file,
            ],
            queue=False,
            api_name=False,
        )
    
        btn.click(
            fn=predict,
            inputs=[
                audio_source,
                input_audio_mic,
                input_audio_file,
                target_language,
            ],
            outputs=[output_text],
            api_name="run",
        )

demo.launch()