Add files via upload

Author: mark-rez

main.py

@@ -0,0 +1,37 @@
+import argparse
+
+# the script in the directory
+import tiktokvoice
+
+def main():
+    # adding arguments
+    parser = argparse.ArgumentParser(description='TikTok TTS')
+    parser.add_argument('-t', help='text input')
+    parser.add_argument('-v', help='voice selection', choices=tiktokvoice.VOICES)
+    parser.add_argument('-n', help='output filename', default='output.mp3')
+    parser.add_argument('-txt', help='text input from a txt file', type=argparse.FileType('r'))
+    parser.add_argument('-play', help='play sound after generating audio', action='store_true')
+
+    args = parser.parse_args()
+
+    # checking if given values are valid
+    if not args.t and not args.txt:
+        print("Insert a valid text or txt file")
+        return
+
+    if args.t and args.txt:
+        print("Only one input type is possible")
+        return
+    
+    if not args.v:
+        print("No voice has been selected")
+        return
+
+    # executing script
+    if args.t:
+        tiktokvoice.tts(args.t, args.v, args.n, args.play)
+    elif args.txt:
+        tiktokvoice.tts(args.txt.read(), args.v, args.n, args.play)
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,2 @@
+requests
+playsound

samples/br_001.mp3

@@ -0,0 +1,175 @@
+import threading, requests, base64
+from playsound import playsound
+
+VOICES = [
+    # DISNEY VOICES
+    'en_us_ghostface',            # Ghost Face
+    'en_us_chewbacca',            # Chewbacca
+    'en_us_c3po',                 # C3PO
+    'en_us_stitch',               # Stitch
+    'en_us_stormtrooper',         # Stormtrooper
+    'en_us_rocket',               # Rocket
+
+    # ENGLISH VOICES
+    'en_au_001',                  # English AU - Female
+    'en_au_002',                  # English AU - Male
+    'en_uk_001',                  # English UK - Male 1
+    'en_uk_003',                  # English UK - Male 2
+    'en_us_001',                  # English US - Female (Int. 1)
+    'en_us_002',                  # English US - Female (Int. 2)
+    'en_us_006',                  # English US - Male 1
+    'en_us_007',                  # English US - Male 2
+    'en_us_009',                  # English US - Male 3
+    'en_us_010',                  # English US - Male 4
+
+    # EUROPE VOICES
+    'fr_001',                     # French - Male 1
+    'fr_002',                     # French - Male 2
+    'de_001',                     # German - Female
+    'de_002',                     # German - Male
+    'es_002',                     # Spanish - Male
+
+    # AMERICA VOICES
+    'es_mx_002',                  # Spanish MX - Male
+    'br_001',                     # Portuguese BR - Female 1
+    'br_003',                     # Portuguese BR - Female 2
+    'br_004',                     # Portuguese BR - Female 3
+    'br_005',                     # Portuguese BR - Male
+
+    # ASIA VOICES
+    'id_001',                     # Indonesian - Female
+    'jp_001',                     # Japanese - Female 1
+    'jp_003',                     # Japanese - Female 2
+    'jp_005',                     # Japanese - Female 3
+    'jp_006',                     # Japanese - Male
+    'kr_002',                     # Korean - Male 1
+    'kr_003',                     # Korean - Female
+    'kr_004',                     # Korean - Male 2
+
+    # SINGING VOICES
+    'en_female_f08_salut_damour',  # Alto
+    'en_male_m03_lobby',           # Tenor
+    'en_female_f08_warmy_breeze',  # Warmy Breeze
+    'en_male_m03_sunshine_soon',   # Sunshine Soon
+
+    # OTHER
+    'en_male_narration',           # narrator
+    'en_male_funny',               # wacky
+    'en_female_emotional',         # peaceful
+]
+
+ENDPOINT = 'https://tiktok-tts.weilnet.workers.dev'
+
+# in one conversion, the text can have a maximum length of 300 characters
+TEXT_BYTE_LIMIT = 300
+
+# create a list by splitting a string, every element has n chars
+def split_string(string: str, chunk_size: int) -> list[str]:
+    words = string.split()
+    result = []
+    current_chunk = ''
+    for word in words:
+        if len(current_chunk) + len(word) + 1 <= chunk_size:  # Check if adding the word exceeds the chunk size
+            current_chunk += ' ' + word
+        else:
+            if current_chunk:  # Append the current chunk if not empty
+                result.append(current_chunk.strip())
+            current_chunk = word
+    if current_chunk:  # Append the last chunk if not empty
+        result.append(current_chunk.strip())
+    return result
+
+# checking if the website that provides the service is available
+def get_api_response() -> requests.Response:
+    url = f'{ENDPOINT}/api/status'
+    response = requests.get(url)
+    return response
+
+# saving the audio file
+def save_audio_file(base64_data: str, filename: str = "output.mp3") -> None:
+    audio_bytes = base64.b64decode(base64_data)
+    with open(filename, "wb") as file:
+        file.write(audio_bytes)
+
+# send POST request to get the audio data
+def generate_audio(text: str, voice: str) -> bytes:
+    url = f'{ENDPOINT}/api/generation'
+    headers = {'Content-Type': 'application/json'}
+    data = {'text': text, 'voice': voice}
+    response = requests.post(url, headers=headers, json=data)
+    return response.content
+
+# creates an text to speech audio file
+def tts(text: str, voice: str = "none", filename: str = "output.mp3", play_sound: bool = False) -> None:
+    # checking if the website is available
+    api_response = get_api_response()
+
+    if api_response.status_code == 200:
+        print("Service available!")
+    else:
+        print("Service not available, try again later or check, if https://tiktok-tts.weilnet.workers.dev is available...")
+        return
+    
+    # checking if arguments are valid
+    if voice == "none":
+        print("No voice has been selected")
+        return
+    
+    if not voice in VOICES:
+        print("Voice does not exist")
+        return
+
+    if len(text) == 0:
+        print("Insert a valid text")
+        return
+
+    # creating the audio file
+    try:
+        if len(text) < TEXT_BYTE_LIMIT:
+            if len(text) < TEXT_BYTE_LIMIT:
+                audio = generate_audio((text), voice)
+                audio_base64_data = str(audio).split('"')[5]
+                
+                if audio_base64_data == "error":
+                    print("This voice is unavailable right now")
+                    return
+                
+        else:
+            # Split longer text into smaller parts
+            text_parts = split_string(text, 299)
+            audio_base64_data = [None] * len(text_parts)
+            
+            # Define a thread function to generate audio for each text part
+            def generate_audio_thread(text_part, index):
+                audio = generate_audio(text_part, voice)
+                base64_data = str(audio).split('"')[5]
+
+                if audio_base64_data == "error":
+                    print("This voice is unavailable right now")
+                    return "error"
+            
+                audio_base64_data[index] = base64_data
+
+            threads = []
+            for index, text_part in enumerate(text_parts):
+                # Create and start a new thread for each text part
+                thread = threading.Thread(target=generate_audio_thread, args=(text_part, index))
+                thread.start()
+                threads.append(thread)
+
+            # Wait for all threads to complete
+            for thread in threads:
+                thread.join()
+                if (thread.result) == "error":
+                    return
+
+            # Concatenate the base64 data in the correct order
+            audio_base64_data = "".join(audio_base64_data)
+            
+        save_audio_file(audio_base64_data, filename)
+        print(f"Audio file saved successfully as '{filename}'")
+        if play_sound:
+            playsound(filename)
+
+    except Exception as e:
+        print("Error occurred while generating audio:", str(e))