Preprocess Module

audiovisually.preprocess.mp3_to_text(audio_path, api_key=None, engine='assemblyai', whisper_model='base', timestamps=False)

Transcribe an audio file to text using AssemblyAI or OpenAI Whisper.

Parameters:

  • audio_path (str) – Path to the audio file.

  • api_key (str, optional) – AssemblyAI API key (required if engine=’assemblyai’).

  • engine (str) – ‘assemblyai’ or ‘whisper’ (default ‘assemblyai’).

  • whisper_model (str) – Whisper model size (default ‘base’).

  • timestamps (bool) – If True and engine=’whisper’, return timestamps.

Returns:

DataFrame with sentences (and optionally timestamps) or error message if transcription fails.

Return type:

pd.DataFrame or str

Example

>>> from audiovisually.preprocess import mp3_to_text
>>> df = mp3_to_text("audio.mp3", api_key="your_api_key", engine="assemblyai")
>>> df = mp3_to_text("audio.mp3", engine="whisper", timestamps=True)
audiovisually.preprocess.translate_data(data, source_lang='auto', dest_lang='en', text_column='Sentence', translated_column='Translation')

Translate text in a DataFrame column, a list of strings, or a single string to a target language.

Parameters:

  • data (pd.DataFrame, list, or str) – DataFrame, list of strings, or single string to translate.

  • source_lang (str) – Source language code (default ‘auto’).

  • dest_lang (str) – Destination language code (default ‘en’).

  • text_column (str) – Name of the column with text to translate (used if data is DataFrame).

  • translated_column (str) – Name of the column to store translations (used if data is DataFrame).

Returns:

DataFrame with translations, translated string, or list of translations.

Return type:

pd.DataFrame, str, or list

Example

>>> from audiovisually.preprocess import translate_data
>>> df = pd.DataFrame({"Sentence": ["Hola", "Bonjour"]})
>>> translated_df = translate_data(df, dest_lang='en')
>>> translated_str = translate_data("Guten Morgen", dest_lang='en')
>>> translated_list = translate_data(["Ciao", "Hallo"], dest_lang='en')
audiovisually.preprocess.video_to_mp3(video_path, output_path=None)

Convert a video file to MP3 audio.

Parameters:

  • video_path (str) – Path to the input video file.

  • output_path (str, optional) – Path to save the output MP3 file. If None, saves in the same folder as input.

Returns:

Path to the generated MP3 file, or error message if conversion fails.

Return type:

str

Example

>>> from audiovisually.preprocess import video_to_mp3
>>> mp3_path = video_to_mp3("input.mp4")