DESCR = """ # TTS Arena Vote on different speech synthesis models! """.strip() INSTR = """ ## Instructions * Listen to two anonymous models * Vote on which one is more natural and realistic * If there's a tie, click Skip *IMPORTANT: Do not only rank the outputs based on naturalness. Also rank based on intelligibility (can you actually tell what they're saying?) and other factors (does it sound like a human?).* **When you're ready to begin, click the Start button below!** The model names will be revealed once you vote. """.strip() LDESC = """ ## Leaderboard A list of the models, based on how highly they are ranked! """.strip() import gradio as gr import random import os import shutil import pandas as pd import sqlite3 from datasets import load_dataset import threading import time from huggingface_hub import HfApi dataset = load_dataset("ttseval/tts-arena", token=os.getenv('HF_TOKEN')) theme = gr.themes.Base( font=[gr.themes.GoogleFont('Libre Franklin'), gr.themes.GoogleFont('Public Sans'), 'system-ui', 'sans-serif'], ) model_names = { 'styletts2': 'StyleTTS 2', 'tacotron': 'Tacotron', 'tacotronph': 'Tacotron Phoneme', 'tacotrondca': 'Tacotron DCA', 'speedyspeech': 'Speedy Speech', 'overflow': 'Overflow TTS', 'vits': 'VITS', 'vitsneon': 'VITS Neon', 'neuralhmm': 'Neural HMM', 'glow': 'Glow TTS', 'fastpitch': 'FastPitch', 'jenny': 'Jenny', 'tortoise': 'Tortoise TTS', 'xtts2': 'XTTSv2', 'xtts': 'XTTS', 'elevenlabs': 'ElevenLabs', 'speecht5': 'SpeechT5', } def get_random_split(existing_split=None): choice = random.choice(list(dataset.keys())) if existing_split and choice == existing_split: return get_random_split(choice) else: return choice def get_db(): return sqlite3.connect('database.db') def create_db(): conn = get_db() cursor = conn.cursor() cursor.execute(''' CREATE TABLE IF NOT EXISTS model ( name TEXT UNIQUE, upvote INTEGER, downvote INTEGER ); ''') def get_data(): conn = get_db() cursor = conn.cursor() cursor.execute('SELECT name, upvote, downvote FROM model') data = cursor.fetchall() df = pd.DataFrame(data, columns=['name', 'upvote', 'downvote']) df['name'] = df['name'].replace(model_names) df['votes'] = df['upvote'] + df['downvote'] # df['score'] = round((df['upvote'] / df['votes']) * 100, 2) # Percentage score ## ELO SCORE df['score'] = 1200 for i in range(len(df)): for j in range(len(df)): if i != j: expected_a = 1 / (1 + 10 ** ((df['score'][j] - df['score'][i]) / 400)) expected_b = 1 / (1 + 10 ** ((df['score'][i] - df['score'][j]) / 400)) actual_a = df['upvote'][i] / df['votes'][i] actual_b = df['upvote'][j] / df['votes'][j] df.at[i, 'score'] += 32 * (actual_a - expected_a) df.at[j, 'score'] += 32 * (actual_b - expected_b) if df['votes'][j] < 3: df.at[j, 'score'] -= (3 - df['votes'][j]) * 5 df['score'] = round(df['score']) ## ELO SCORE df = df.sort_values(by='score', ascending=False) # df = df[['name', 'score', 'upvote', 'votes']] df = df[['name', 'score', 'votes']] return df def get_random_splits(): choice1 = get_random_split() choice2 = get_random_split(choice1) return (choice1, choice2) def upvote_model(model): conn = get_db() cursor = conn.cursor() cursor.execute('UPDATE model SET upvote = upvote + 1 WHERE name = ?', (model,)) if cursor.rowcount == 0: cursor.execute('INSERT OR REPLACE INTO model (name, upvote, downvote) VALUES (?, 1, 0)', (model,)) conn.commit() cursor.close() def downvote_model(model): conn = get_db() cursor = conn.cursor() cursor.execute('UPDATE model SET downvote = downvote + 1 WHERE name = ?', (model,)) if cursor.rowcount == 0: cursor.execute('INSERT OR REPLACE INTO model (name, upvote, downvote) VALUES (?, 0, 1)', (model,)) conn.commit() cursor.close() def a_is_better(model1, model2): if model1 and model2: upvote_model(model1) downvote_model(model2) return reload(model1, model2) def b_is_better(model1, model2): if model1 and model2: upvote_model(model2) downvote_model(model1) return reload(model1, model2) def both_bad(model1, model2): if model1 and model2: downvote_model(model1) downvote_model(model2) return reload(model1, model2) def both_good(model1, model2): if model1 and model2: upvote_model(model1) upvote_model(model2) return reload(model1, model2) def reload(chosenmodel1=None, chosenmodel2=None): # Select random splits split1, split2 = get_random_splits() d1, d2 = (dataset[split1], dataset[split2]) choice1, choice2 = (d1.shuffle()[0]['audio'], d2.shuffle()[0]['audio']) if chosenmodel1 in model_names: chosenmodel1 = model_names[chosenmodel1] if chosenmodel2 in model_names: chosenmodel2 = model_names[chosenmodel2] out = [ (choice1['sampling_rate'], choice1['array']), (choice2['sampling_rate'], choice2['array']), split1, split2 ] if chosenmodel1: out.append(f'This model was {chosenmodel1}') if chosenmodel2: out.append(f'This model was {chosenmodel2}') return out with gr.Blocks() as leaderboard: gr.Markdown(LDESC) # df = gr.Dataframe(interactive=False, value=get_data()) df = gr.Dataframe(interactive=False) leaderboard.load(get_data, outputs=[df]) with gr.Blocks() as vote: gr.Markdown(INSTR) with gr.Row(): gr.HTML('

Model A

') gr.HTML('

Model B

') model1 = gr.Textbox(interactive=False, visible=False) model2 = gr.Textbox(interactive=False, visible=False) with gr.Group(): with gr.Row(): prevmodel1 = gr.Textbox(interactive=False, show_label=False, container=False, value="Vote to reveal model A") prevmodel2 = gr.Textbox(interactive=False, show_label=False, container=False, value="Vote to reveal model B", text_align="right") with gr.Row(): aud1 = gr.Audio(interactive=False, show_label=False, show_download_button=False, show_share_button=False, waveform_options={'waveform_progress_color': '#3C82F6'}) aud2 = gr.Audio(interactive=False, show_label=False, show_download_button=False, show_share_button=False, waveform_options={'waveform_progress_color': '#3C82F6'}) with gr.Row(): abetter = gr.Button("A is Better", variant='primary') bbetter = gr.Button("B is Better", variant='primary') with gr.Row(): bothbad = gr.Button("Both are Bad", scale=2) skipbtn = gr.Button("Skip", scale=1) bothgood = gr.Button("Both are Good", scale=2) outputs = [aud1, aud2, model1, model2, prevmodel1, prevmodel2] abetter.click(a_is_better, outputs=outputs, inputs=[model1, model2]) bbetter.click(b_is_better, outputs=outputs, inputs=[model1, model2]) skipbtn.click(b_is_better, outputs=outputs, inputs=[model1, model2]) bothbad.click(both_bad, outputs=outputs, inputs=[model1, model2]) bothgood.click(both_good, outputs=outputs, inputs=[model1, model2]) vote.load(reload, outputs=[aud1, aud2, model1, model2]) with gr.Blocks(theme=theme, css="footer {visibility: hidden}", title="TTS Leaderboard") as demo: gr.Markdown(DESCR) gr.TabbedInterface([vote, leaderboard], ['Vote', 'Leaderboard']) def restart_space(): api = HfApi( token=os.getenv('HF_TOKEN') ) time.sleep(60 * 60) # Every hour print("Restarting space") api.restart_space(repo_id=os.getenv('HF_ID')) def sync_db(): api = HfApi( token=os.getenv('HF_TOKEN') ) while True: time.sleep(60 * 5) print("Uploading DB") api.upload_file( path_or_fileobj='database.db', path_in_repo='database.db', repo_id=os.getenv('DATASET_ID'), repo_type='dataset' ) if os.getenv('HF_ID'): restart_thread = threading.Thread(target=restart_space) restart_thread.daemon = True restart_thread.start() if os.getenv('DATASET_ID'): # Fetch DB api = HfApi( token=os.getenv('HF_TOKEN') ) print("Downloading DB...") try: path = api.hf_hub_download( repo_id=os.getenv('DATASET_ID'), repo_type='dataset', filename='database.db', cache_dir='./' ) shutil.copyfile(path, 'database.db') print("Downloaded DB") except: pass # Update DB db_thread = threading.Thread(target=sync_db) db_thread.daemon = True db_thread.start() create_db() demo.queue(api_open=False).launch(show_api=False)