{"id":317,"date":"2024-03-30T14:06:19","date_gmt":"2024-03-30T14:06:19","guid":{"rendered":"https:\/\/noerguerra.com\/?p=317"},"modified":"2024-10-26T06:05:55","modified_gmt":"2024-10-26T06:05:55","slug":"how-to-read-text-aloud-with-piper-and-python","status":"publish","type":"post","link":"https:\/\/noerguerra.com\/blog\/how-to-read-text-aloud-with-piper-and-python\/","title":{"rendered":"How to read text aloud with Piper and Python"},"content":{"rendered":"\n

Piper<\/strong> is a neural text-to-speech system that can run locally and deliver great sounding audio clips even on underpowered computers. Piper is optimized to run on the Raspberry Pi 4, and you can easily import it to your application as a library.<\/p>\n\n\n\n

I stumbled upon Piper TTS after looking for a simple text-to-speech application where I could input text and read it out loud on Linux Minut. Coming from using Microsoft online voices on Microsoft Edge, I was looking for a more natural sound than the robotic voices from programs like Festival or eSpeak, and I was impressed by the natural sound of Piper and its capabilities to run smoothly on almost any kind of modern computer.<\/p>\n\n\n\n

You can listen to samples generated using Piper here<\/a>.<\/p>\n\n\n\n

Installing Piper on Linux<\/h2>\n\n\n\n

You can install Piper through pip with this command:<\/p>\n\n\n\n

pip install piper-tts<\/code><\/pre>\n\n\n\n
Afterwards, you should be able to import piper<\/code> into your program with import piper<\/code>.<\/p>\n\n\n\n
You can quickly test piper from the terminal by piping the output of a program for it to read, like this:<\/p>\n\n\n\n
echo \"Hello world! This is text to speech\" | piper \\\n--model en_US-lessac-medium \\\n--output_file audio.wav<\/code><\/pre>\n\n\n\n
The resulting audio will be saved in audio.wav<\/code> and can be played with any media player.<\/p>\n\n\n\n
Adding models to Piper<\/h2>\n\n\n\n
The Piper repository includes a variety of pre-trained voice models<\/a> sorted by language that you can use in your projects. These models determine how the synthesized speech will sound, in other words, each model is a different “voice” you can use with Piper (Although occasionally a model will contain multiple voices). <\/p>\n\n\n\n
You can also create your own voice model using the Piper Recording Studio, a web application that you can run locally to generate a Piper dataset by recording clips with your voice. However, make sure to have a decent graphics card on your device, or training your model could be a very slow task. For more information on the process of creating a model for Piper, look at this article from Sam Howell<\/a>.<\/p>\n\n\n\n
In order to add models to Piper, you need to obtain the .onnx<\/code> format and the .onnx.json<\/code> file. These JSON files contain important metadata about the models, such as their sample rate and phoneme set, and must always have the same name as the .onnx<\/code> file, and be located within the same directory.<\/p>\n\n\n\n
For example:<\/p>\n\n\n\n
directory\/\n|-- ...\n|-- es_MX-claude-high.onnx\n|-- es_MX-claude-high.onnx.json\n<\/code><\/pre>\n\n\n\n
Generating audio files from text<\/h2>\n\n\n\n
To generate audio files programatically using Piper, you’ll need to import the PiperVoice<\/code> class and use the appropiate methods, like this (based on this answer<\/a>):<\/p>\n\n\n\n
import wave\nfrom piper.voice import PiperVoice\n\nmodel = \"\/path\/to\/model.onnx\"\nvoice = PiperVoice.load(model)\ntext = \"This is an example of text to speech\"\nwav_file = wave.open(\"output.wav\", \"w\")\naudio = voice.synthesize(text, wav_file)<\/code><\/pre>\n\n\n\n
In this code, we:<\/p>\n\n\n\n