Text-to-Speech
English

How to generate single audio file for longer texts.

#87
by AgarVaibhav - opened

In this version seperate audio files are being created as the code is written, but is there any other way through which i can a single audio file for longer text?

Yeah it wouldn't be that hard, I made this for XTTS:

def concatenate_wav_files(self, output_files):
    # Vine each grape into bunch upon completion
    combined_audio = AudioSegment.empty()
    for file in output_files:
        #file = resemble_enchance_audio(file, True)
        audio_segment = AudioSegment.from_wav(file)
        combined_audio += audio_segment
        os.remove(file)  # Currently not working(?) in original venv

    # Save the combined audio file
    output_dir = Path.cwd() / "tts_outputs"
    output_dir.mkdir(exist_ok=True)
    save_path, _ = QFileDialog.getSaveFileName(self, "Save Audio File", str(output_dir), "Wave Files (*.wav)")
    
    if save_path:
        combined_audio.export(save_path, format="wav")
        print(f"Combined audio saved to: {save_path}")

I'm sure something already exists for this, if not I'll send you whatever I come up with.

Sign up or log in to comment