tts_arabic

Arabisches TTS-Modell (Fastpitch) aus der TTS-Arabic-Pytorch Repow im ONNX-Format.

Audio -Samples finden Sie hier.

Installieren mit:

 pip install git+https://github.com/nipponjo/tts_arabic.git

Beispiele:

 # %%
from tts_arabic import tts

# %%
text = "اَلسَّلامُ عَلَيكُم يَا صَدِيقِي."
wave = tts ( text , speaker = 2 , pace = 0.9 , play = True )

# %% Buckwalter transliteration
text = ">als~alAmu Ealaykum yA Sadiyqiy."
wave = tts ( text , speaker = 0 , play = True )

# %% Unvocalized input
text_unvoc = "القهوة مشروب يعد من بذور البن المحمصة"
wave = tts ( text_unvoc , play = True , vowelizer = 'shakkelha' )

Vorbereitete Modelle:

Modell	Modell ID	Typ	#Params	Papier
Fastpitch	Fastpitch	Text-> mel	46,3 m	Arxiv
Mixrtts	Mixer128	Text-> mel	2,9 m	Arxiv
Mixrtts	Mixer80	Text-> mel	1,5 m	Arxiv
Hifi-Gan	HiFigan	Vocoder	13,9 m	Arxiv
Vokos	Vokos	Vocoder	13,4 m	Arxiv

TTS -Optionen:

 from tts_arabic import tts

text = "اَلسَّلامُ عَلَيكُم يَا صَدِيقِي."
wave = tts (
    text , # input text
    speaker = 1 , # speaker id; choose between 0,1,2,3
    pace = 1 , # speaker pace
    denoise = 0.005 , # vocoder denoiser strength
    play = True , # play audio?
    pitch_mul = 1 , # pitch multiplier
    pitch_add = 0 , # pitch offset
    vowelizer = None , # vowelizer model
    model_id = 'fastpitch' , # Model ID for Text->Mel model
    vocoder_id = 'hifigan' , # Model ID for vocoder model
    cuda = None , # Optional; CUDA device index
    save_to = './test.wav' , # Optionally; save audio WAV file
    bits_per_sample = 32 , # when save_to is specified (8, 16 or 32 bits)
    )