app.py

import spaces
import gradio as gr
import torch
from transformers.models.speecht5.number_normalizer import EnglishNumberNormalizer
from string import punctuation
import re

from parler_tts import ParlerTTSForConditionalGeneration
from transformers import AutoTokenizer, AutoFeatureExtractor, set_seed

device = "cuda:0" if torch.cuda.is_available() else "cpu"

repo_id = "parler-tts/parler-tts-mini-expresso"

model = ParlerTTSForConditionalGeneration.from_pretrained(repo_id).to(device)
tokenizer = AutoTokenizer.from_pretrained(repo_id)
feature_extractor = AutoFeatureExtractor.from_pretrained(repo_id)

SAMPLE_RATE = feature_extractor.sampling_rate
SEED = 42

default_text = "*Remember* - this is only the first iteration of the model! To improve the prosody and naturalness of the speech further, we're scaling up the amount of training data by a factor of *five times*."
default_description = "Thomas speaks with emphasis and excitement at a moderate pace with high quality."
examples = [
    [
        "Remember - this is only the first iteration of the model. To improve the prosody and naturalness of the speech further, we're scaling up the amount of training data by a factor of five times.",
        "Thomas speaks in a sad tone at a very slow pace with high quality."
    ],
    [
        "Did you know? <laugh> You can reproduce this entire training recipe by following the steps outlined on the model card!",
        "Talia speaks quickly with laughter and high quality audio.",
    ],
    [
        "But that's no secret! The entire project is open-source first, with all release artefacts on the Hub.",
        "Elisabeth speaks happily at a slightly slower than average pace with high quality audio.",
    ],
    [
        "Hey there. I'm Jerry. Or at least, I *think* I am? I just need to check that quickly.",
        "Jerry speaks in a confused tone at a moderate pace with high quality audio.",
    ],
]

number_normalizer = EnglishNumberNormalizer()


def preprocess(text):
    text = number_normalizer(text).strip()
    text = text.replace("-", " ")
    if text[-1] not in punctuation:
        text = f"{text}."

    abbreviations_pattern = r'\b[A-Z][A-Z\.]+\b'

    def separate_abb(chunk):
        chunk = chunk.replace(".", "")
        print(chunk)
        return " ".join(chunk)

    abbreviations = re.findall(abbreviations_pattern, text)
    for abv in abbreviations:
        if abv in text:
            text = text.replace(abv, separate_abb(abv))
    return text


@spaces.GPU
def gen_tts(text, description):
    inputs = tokenizer(description, return_tensors="pt").to(device)
    prompt = tokenizer(preprocess(text), return_tensors="pt").to(device)

    set_seed(SEED)
    generation = model.generate(
        input_ids=inputs.input_ids, prompt_input_ids=prompt.input_ids, do_sample=True, temperature=1.0
    )
    audio_arr = generation.cpu().numpy().squeeze()

    return SAMPLE_RATE, audio_arr


css = """
        #share-btn-container {
            display: flex;
            padding-left: 0.5rem !important;
            padding-right: 0.5rem !important;
            background-color: #000000;
            justify-content: center;
            align-items: center;
            border-radius: 9999px !important; 
            width: 13rem;
            margin-top: 10px;
            margin-left: auto;
            flex: unset !important;
        }
        #share-btn {
            all: initial;
            color: #ffffff;
            font-weight: 600;
            cursor: pointer;
            font-family: 'IBM Plex Sans', sans-serif;
            margin-left: 0.5rem !important;
            padding-top: 0.25rem !important;
            padding-bottom: 0.25rem !important;
            right:0;
        }
        #share-btn * {
            all: unset !important;
        }
        #share-btn-container div:nth-child(-n+2){
            width: auto !important;
            min-height: 0px !important;
        }
        #share-btn-container .wrap {
            display: none !important;
        }
"""
with gr.Blocks(css=css) as block:
    gr.HTML(
        """
            <div style="text-align: center; max-width: 700px; margin: 0 auto;">
              <div
                style="
                  display: inline-flex; align-items: center; gap: 0.8rem; font-size: 1.75rem;
                "
              >
                <h1 style="font-weight: 900; margin-bottom: 7px; line-height: normal;">
                  Parler-TTS: Expresso ☕️️
                </h1>
              </div>
            </div>
        """
    )
    gr.HTML(
        f"""
        <p><a href="https://huggingface.co/parler-tts/parler-tts-mini-expresso"> Parler-TTS Mini: Expresso</a>
        is a text-to-speech (TTS) model fine-tuned on the <a href="https://huggingface.co/datasets/ylacombe/expresso"> Expresso dataset</a>.
        It generates high-quality speech in a given <b>emotion</b> and <b>voice</b> that can be controlled through a simple text prompt.</p>

        <p>Tips for ensuring good generation:
        <ul>
            <li>Specify the name of a male speaker (Jerry, Thomas) or female speaker (Talia, Elisabeth) for consistent voices</li>
            <li>The model can generate in a range of emotions, including: "happy", "confused", "default" (meaning no particular emotion conveyed), "laughing", "sad", "whisper", "emphasis"</li>
            <li>Punctuation can be used to control the prosody of the generations, e.g. use commas to add small breaks in speech</li>
            <li>Wrap words in asterisk to emphasise them (e.g. `*Remember*` in the example below)</li>
        </ul>
        </p>
        """
    )
    with gr.Row():
        with gr.Column():
            input_text = gr.Textbox(label="Input Text", lines=2, value=default_text, elem_id="input_text")
            description = gr.Textbox(label="Description", lines=2, value=default_description, elem_id="input_description")
            run_button = gr.Button("Generate Audio", variant="primary")
        with gr.Column():
            audio_out = gr.Audio(label="Parler-TTS generation", type="numpy", elem_id="audio_out")

    inputs = [input_text, description]
    outputs = [audio_out]
    gr.Examples(examples=examples, fn=gen_tts, inputs=inputs, outputs=outputs, cache_examples=True)
    run_button.click(fn=gen_tts, inputs=inputs, outputs=outputs, queue=True)
    gr.HTML(
        """
        <p>To improve the prosody and naturalness of the speech further, we're scaling up the amount of training data to 50k hours of speech.
        The v1 release of the model will be trained on this data, as well as inference optimisations, such as flash attention
        and torch compile, that will improve the latency by 2-4x. If you want to find out more about how this model was trained and even fine-tune it yourself, check-out the 
        <a href="https://github.com/huggingface/parler-tts"> Parler-TTS</a> repository on GitHub. The Parler-TTS codebase and its associated checkpoints are licensed under <a href='https://github.com/huggingface/parler-tts?tab=Apache-2.0-1-ov-file#readme'> Apache 2.0</a>.</p>
        """
    )

block.queue()
block.launch(share=True)