------------- THIS VERSION OF TTS IS DISCONTINUED --------- PLEASE REFER TO UPDATED VERSION HERE: https://github.com/OlafKranse/coqui-ai_tts_for_documents Tested on Ubuntu 20.0.4 subsystem for windows, using python 3.8 . The script uses Neural TTS by mozilla; https://github.com/mozilla/TTS Find installation guide here: https://github.com/mozilla/TTS/wiki/TTS-Notebooks-and-Tutorials ; English - LJSpeech + MultiBandMelGAN + Tacotorn2(DDC): On CPU real-time speech synthesis with PyTorch Overal most of the script they use is the same code; but there is a loop included that takes the input file and synthesizes sentece by sentece. Another script is then used to merge these .wav files into one big wav file. Here and there, even though rare, the script does a hiccup or gets stuck, it will then usually skip a sentence. ### Setup ### 1. Clone the git https://github.com/mozilla/TTS; or use the notebook mentioned above to install the script. 2. Copy the script "synth_long_for_batch.py" into the main folder where you have your config.json. 3. Make a folder batch, containing the following folders: in, out . in this folder copy the script "merging_wavs.py". 4. Fill the in folder with as many text files as you want to synthesize as a single line text file per text you want to synth. 5. run the script "synth_long_for_batch.py", this will now synth sentence for sentence all your files in the "in" folder 6. after the synthesizing is done, run the script "merging_wavs.py" . the merged files are in the out folder.
Olaf2K/Neural_TTS
This is a custom python script using Mozilla neural TTS that takes a batch of input files as text, and turns them into one long .wav file.
Python