app.py

import gradio as gr

#Get models
#ASR model for input speech
speech2text = gr.Interface.load("huggingface/facebook/wav2vec2-base-960h",
                                inputs=gr.inputs.Audio(label="Record Audio File", type="file", source = "microphone"))
                                        
#translates english to spanish text                      
translator = gr.Interface.load("huggingface/Helsinki-NLP/opus-mt-en-es",
                                outputs=gr.outputs.Textbox(label="English to Spanish Translated Text"))
#TTS model for output speech                                
text2speech = gr.Interface.load("huggingface/facebook/tts_transformer-es-css10",
                                outputs=gr.outputs.Audio(label="English to Spanish Translated Audio"),
                                allow_flagging="never")

                                
translate = gr.Series(speech2text, translator) #outputs Spanish text translation
en2es = gr.Series(translate, text2speech) #outputs Spanish audio
ui = gr.Parallel(translate, en2es) #allows transcription of Spanish audio

#gradio interface
ui.title = "English to Spanish Speech Translator"
ui.description = """<center>A useful tool in translating English to Spanish audio. All pre-trained models are found in huggingface.</center>"""
ui.examples = [['ljspeech.wav'],['ljspeech2.wav',]]
ui.theme = "peach"
ui.article = article=""<h2>Pre-trained model Information</h2>
                       <h3>Automatic Speech Recognition</h3>
                       <p style='text-align: justify'>The model used for the ASR part of this space is from                
                       <https://huggingface.co/facebook/wav2vec2-base-960h> which is pretrained and fine-tuned on <b>960 hours of 
                       Librispeech</b> on 16kHz sampled speech audio. This model has a <b>word error rate (WER)</b> of <b>8.6 percent on 
                       noisy speech</b> and <b>5.2 percent on clean speech</b> on the standard LibriSpeech benchmark. More information can be 
                       found on its website at <https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/> and 
                       original model is under <https://github.com/pytorch/fairseq/tree/main/examples/wav2vec>.</p>
                       <h3>Text Translator</h3>
                       <p style='text-align: justify'>The English to Spanish text translator pre-trained model is from <Helsinki-NLP/opus-
                       mt-en-es> which is part of the <b>The Tatoeba Translation Challenge (v2021-08-07)</b> as seen from its github repo at 
                       <https://github.com/Helsinki-NLP/Tatoeba-Challenge>. This project aims to develop machine translation in real-world 
                       cases for many languages. </p>
                       <h3>Text to Speech</h3>
                       <p style='text-align: justify'> The TTS model used is from <https://huggingface.co/facebook/tts_transformer-es-css10>. 
                       This model uses the <b>Fairseq(-py)</b> sequence modeling toolkit for speech synthesis, in this case, specifically TTS 
                       for Spanish. More information can be seen on their git at <https://github.com/pytorch/fairseq>. </p>
                    ""   
                       
                       
ui.launch(inbrowser=True)