Update app.py

app.py

@@ -1,134 +1,86 @@
 import gradio as gr
-import torchaudio
-from datasets import load_dataset, load_metric
-from transformers import (
-    Wav2Vec2ForCTC,
-    Wav2Vec2Processor,
-    AutoTokenizer, 
-    AutoModelWithLMHead 
-)
-import torch
-import re
-import sys
-import soundfile as sf
-
-
-
+from transformers import pipeline
+from pydub import AudioSegment
+import os
+import speech_recognition as sr
 
 model_name = "voidful/wav2vec2-xlsr-multilingual-56"
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-processor_name = "voidful/wav2vec2-xlsr-multilingual-56"
-
-import pickle
-with open("lang_ids.pk", 'rb') as output:
-    lang_ids = pickle.load(output)
-    
-model = Wav2Vec2ForCTC.from_pretrained(model_name).to(device)
-processor = Wav2Vec2Processor.from_pretrained(processor_name)
-
-model.eval()
-
-def load_file_to_data(file,sampling_rate=16_000):
-    batch = {}
-    speech, _ = torchaudio.load(file)
-    if sampling_rate != '16_000' or sampling_rate != '16000':
-        resampler = torchaudio.transforms.Resample(orig_freq=sampling_rate, new_freq=16_000)
-        batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
-        batch["sampling_rate"] = resampler.new_freq
-    else:
-        batch["speech"] = speech.squeeze(0).numpy()
-        batch["sampling_rate"] = '16000'
-    return batch
-
-
-def predict(data):
-    data=load_file_to_data(data,sampling_rate=16_000)
-    features = processor(data["speech"], sampling_rate=data["sampling_rate"], padding=True, return_tensors="pt")
-    input_values = features.input_values.to(device)
-    attention_mask = features.attention_mask.to(device)
-    with torch.no_grad():
-        logits = model(input_values, attention_mask=attention_mask).logits
-        decoded_results = []
-        for logit in logits:
-            pred_ids = torch.argmax(logit, dim=-1)
-            mask = pred_ids.ge(1).unsqueeze(-1).expand(logit.size())
-            vocab_size = logit.size()[-1]
-            voice_prob = torch.nn.functional.softmax((torch.masked_select(logit, mask).view(-1,vocab_size)),dim=-1)
-            comb_pred_ids = torch.argmax(voice_prob, dim=-1)
-            decoded_results.append(processor.decode(comb_pred_ids))
-
-    return decoded_results
-
-def predict_lang_specific(data,lang_code):
-    data=load_file_to_data(data,sampling_rate=16_000)
-    features = processor(data["speech"], sampling_rate=data["sampling_rate"], padding=True, return_tensors="pt")
-    input_values = features.input_values.to(device)
-    attention_mask = features.attention_mask.to(device)
-    with torch.no_grad():
-        logits = model(input_values, attention_mask=attention_mask).logits
-        decoded_results = []
-        for logit in logits:
-            pred_ids = torch.argmax(logit, dim=-1)
-            mask = ~pred_ids.eq(processor.tokenizer.pad_token_id).unsqueeze(-1).expand(logit.size())
-            vocab_size = logit.size()[-1]
-            voice_prob = torch.nn.functional.softmax((torch.masked_select(logit, mask).view(-1,vocab_size)),dim=-1)
-            filtered_input = pred_ids[pred_ids!=processor.tokenizer.pad_token_id].view(1,-1).to(device)
-            if len(filtered_input[0]) == 0:
-                decoded_results.append("")
-            else:
-                lang_mask = torch.empty(voice_prob.shape[-1]).fill_(0)
-                lang_index = torch.tensor(sorted(lang_ids[lang_code]))
-                lang_mask.index_fill_(0, lang_index, 1)
-                lang_mask = lang_mask.to(device)
-                comb_pred_ids = torch.argmax(lang_mask*voice_prob, dim=-1)
-                decoded_results.append(processor.decode(comb_pred_ids))
-                
-    return decoded_results
-
-'''def recognition(audio_file):
-    print("audio_file", audio_file.name)
-    speech, rate = sp.load_speech_with_file(audio_file.name)
+model0 = pipeline(task="automatic-speech-recognition",
+                 model=model_name)
+
+
+model_name = "SLPL/Sharif-wav2vec2"
+model = pipeline(task="automatic-speech-recognition",
+                 model=model_name)
+model_name = "ghofrani/common8"
+model1 = pipeline(task="automatic-speech-recognition",
+                 model=model_name)
+
+import json
+def predict_fa(speech,model):
+    if model== "SLPL/Sharif-wav2vec2":
+        text = model(speech,return_timestamps="word" )
+    else if model== "ghofrani/common8":
+        text = model1(speech,return_timestamps="word" )
+    else if model== "voidful/wav2vec2-xlsr-multilingual-56":
+        text = model0(speech,return_timestamps="word" )
+    return [text['text'],json.dumps(text)]
+
+
+def convert_to_wav(filename):
+    filenameObj=os.path.splitext(filename)
+    audio = AudioSegment.from_file(filename,format=filenameObj[1].replace(".",""))
+    new_filename = filenameObj[0] + ".wav"
+    while os.path.exists(new_filename):
+        new_filename = os.path.splitext(new_filename)[0]+"(1)"+ ".wav"
+    audio.export(new_filename, format="wav")
+    print(f"Converting {filename} to {new_filename}...")
+    return new_filename
+def g_rec(audio_File ,language):
+    r = sr.Recognizer()
     
-    result = sp.predict_audio_file(speech)
-    print(result)
-
-    return result
-'''
+    if not os.path.splitext(audio_File)[1]==".wav":
+        audio_File=convert_to_wav(audio_File)
+    hellow=sr.AudioFile(audio_File)
+    with hellow as source:
+        audio = r.record(source)
+    try:
+        s = r.recognize_google(audio,language =language)
+        res= "Text: "+s
+    except Exception as e:
+        res= "Exception: "+str(e)
+    return res
+  # Export file as .wav
+  
 #predict(load_file_to_data('audio file path',sampling_rate=16_000)) # beware of the audio file sampling rate
 
 #predict_lang_specific(load_file_to_data('audio file path',sampling_rate=16_000),'en') # beware of the audio file sampling rate
 with gr.Blocks() as demo:
     gr.Markdown("multilingual Speech Recognition")
-    with gr.Tab("Auto"):
-        gr.Markdown("automatically detects your language")
-        inputs_speech =gr.Audio(source="upload", type="filepath", optional=True)
-        output_transcribe = gr.HTML(label="")
-        transcribe_audio= gr.Button("Submit")
-    with gr.Tab("manual"):
+
+    with gr.Tab("Persian models"):
+        inputs_speech_fa =gr.Audio(source="upload", type="filepath", optional=True,label="Upload your audio:")
+        inputs_model_fa =gr.inputs.Radio(label="Language", choices=["ghofrani/common8","SLPL/Sharif-wav2vec2","voidful/wav2vec2-xlsr-multilingual-56"])
+        output_transcribe1_fa = gr.Textbox(label="Transcribed text:")
+        output_transcribe1_fa1 = gr.Textbox(label="Transcribed text with timestamps:")
+        transcribe_audio1_fa= gr.Button("Submit")
+    with gr.Tab("google"):
         gr.Markdown("set your speech language")
         inputs_speech1 =[ 
             gr.Audio(source="upload", type="filepath"),
-            gr.Dropdown(choices=["ar","as","br","ca","cnh","cs","cv","cy","de","dv","el","en","eo","es","et","eu","fa","fi","fr","fy-NL","ga-IE","hi","hsb","hu","ia","id","it","ja","ka","ky","lg","lt","lv","mn","mt","nl","or","pa-IN","pl","pt","rm-sursilv","rm-vallader","ro","ru","sah","sl","sv-SE","ta","th","tr","tt","uk","vi","zh-CN","zh-HK","zh-TW"]
-,value="fa",label="language code")
+            gr.Dropdown(choices=["af-ZA","am-ET","ar-AE","ar-BH","ar-DZ","ar-EG","ar-IL","ar-IQ","ar-JO","ar-KW","ar-LB","ar-MA","ar-MR","ar-OM","ar-PS","ar-QA","ar-SA","ar-TN","ar-YE","az-AZ","bg-BG","bn-BD","bn-IN","bs-BA","ca-ES","cs-CZ","da-DK","de-AT","de-CH","de-DE","el-GR","en-AU","en-CA","en-GB","en-GH","en-HK","en-IE","en-IN","en-KE","en-NG","en-NZ","en-PH","en-PK","en-SG","en-TZ","en-US","en-ZA","es-AR","es-BO","es-CL","es-CO","es-CR","es-DO","es-EC","es-ES","es-GT","es-HN","es-MX","es-NI","es-PA","es-PE","es-PR","es-PY","es-SV","es-US","es-UY","es-VE","et-EE","eu-ES","fa-IR","fi-FI","fil-PH","fr-BE","fr-CA","fr-CH","fr-FR","gl-ES","gu-IN","hi-IN","hr-HR","hu-HU","hy-AM","id-ID","is-IS","it-CH","it-IT","iw-IL","ja-JP","jv-ID","ka-GE","kk-KZ","km-KH","kn-IN","ko-KR","lo-LA","lt-LT","lv-LV","mk-MK","ml-IN","mn-MN","mr-IN","ms-MY","my-MM","ne-NP","nl-BE","nl-NL","no-NO","pa-Guru-IN","pl-PL","pt-BR","pt-PT","ro-RO","ru-RU","si-LK","sk-SK","sl-SI","sq-AL","sr-RS","su-ID","sv-SE","sw-KE","sw-TZ","ta-IN","ta-LK","ta-MY","ta-SG","te-IN","th-TH","tr-TR","uk-UA","ur-IN","ur-PK","uz-UZ","vi-VN","yue-Hant-HK","zh (cmn-Hans-CN)","zh-TW (cmn-Hant-TW)","zu-ZA"]
+,value="fa-IR",label="language code")
         ]
         output_transcribe1 = gr.Textbox(label="output")
         transcribe_audio1= gr.Button("Submit")
-    '''with gr.Tab("Auto1"):
-        gr.Markdown("automatically detects your language")
-        inputs_speech2 = gr.Audio(label="Input Audio", type="file")
-        output_transcribe2 = gr.Textbox()
-        transcribe_audio2= gr.Button("Submit")'''
-    transcribe_audio.click(fn=predict,
-    inputs=inputs_speech,
-    outputs=output_transcribe)
     
-    transcribe_audio1.click(fn=predict_lang_specific,
+    transcribe_audio1_fa.click(fn=predict_fa,
+    inputs=[inputs_speech_fa ,inputs_model_fa ],
+    outputs=[output_transcribe1_fa ,output_transcribe1_fa1 ] )
+    
+    transcribe_audio1_go.click(fn=g_rec,
     inputs=inputs_speech1 ,
     outputs=output_transcribe1 )
-    
-    '''transcribe_audio2.click(fn=recognition,
-    inputs=inputs_speech2 ,
-    outputs=output_transcribe2 )'''
 
 
 if __name__ == "__main__":