testing subtitles

README.md

@@ -41,4 +41,9 @@ The user will logging using a password and user specified by me. That user and p
 - [ ] Add mel spectrogram?
 - [ ] Add Whisper parameters to the interface
 - [x] Add Whisper X  
-- [ ] Introduce SRT as output
+- [ ] Introduce SRT as output
+- [ ] Obtain txt with Diarization.
+- [ ] Obtain plain txt with segments. 
+- [ ] Introduce POS
+
+Introducir segmento en la app para hacer analisis POS. Quizas correcciones. 

app.py

@@ -55,30 +55,30 @@ def transcribeWhisperX(audiofile, model, language, patiente,
     #Archivo
     #nombre_archivo = guardar_en_archivo(out)
 
-    # ##########################################################################
-    # import whisperx
-    # from pathlib import Path
-    # device = "cuda"
-    # model_a, metadata = whisperx.load_align_model(
-    #     language_code="es", device=device
-    # )
-    # result_aligned = whisperx.align(
-    #     result["segments"],
-    #     model_a,
-    #     metadata,
-    #     vocal_path,
-    #     device=device,
-    # )
-    # import datetime
-    # fecha_actual = datetime.datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
+    ##########################################################################
+    import whisperx
+    from pathlib import Path
+    device = "cuda"
+    model_a, metadata = whisperx.load_align_model(
+        language_code="es", device=device
+    )
+    result_aligned = whisperx.align(
+        result["segments"],
+        model_a,
+        metadata,
+        vocal_path,
+        device=device,
+    )
+    import datetime
+    fecha_actual = datetime.datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
     
-    # # Generar el nombre del archivo
-    # nombre_archivo = f"transcription_{fecha_actual}.srt"
+    # Generar el nombre del archivo
+    nombre_archivo = f"transcription_{fecha_actual}.srt"
 
-    # file_path = Path(nombre_archivo)
-    # writter_args = {"highlight_words": None, "max_line_count": None, "max_line_width": None}
-    # srt_writer = get_writer("srt", Path("."))
-    # srt_writer(result_aligned["segments"], str(file_path.stem), writter_args)
+    file_path = Path(nombre_archivo)
+    writter_args = {"highlight_words": None, "max_line_count": None, "max_line_width": None}
+    srt_writer = get_writer("srt", Path("."))
+    srt_writer(result_aligned["segments"], str(file_path.stem), writter_args)
 
     # with open(
     #    nombre_archivo,
@@ -86,7 +86,7 @@ def transcribeWhisperX(audiofile, model, language, patiente,
     #     encoding="utf-8",
     # ) as srt:
     #     write_srt(result_aligned["segments"], file=srt)
-    ############################################################################
+    ###########################################################################
 
     return audio_path, audio_normalized_path, vocal_path, novocal_path, vocal_path, guardar_dataframe_en_csv(diarize_segments), json.dumps(result)