Edit model card

This model is a fine-tuned version of SpeechT5 for the Ukrainian language, using the Common Voice dataset.

Usage:

!pip install git+https://github.com/huggingface/transformers

from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech, SpeechT5ForSpeechToSpeech, SpeechT5HifiGan

processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
model = SpeechT5ForTextToSpeech.from_pretrained("tenebrisu/speecht5_tts_common_voice_uk")
vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")

speaker_embeddings = 2 * torch.rand((1, 512)) - 1
text = """ pryvit yak spravy """

inputs = processor(text=text, return_tensors="pt")

speech = model.generate_speech(inputs["input_ids"], speaker_embeddings, output_cross_attentions=True)
waveform = vocoder.forward(speech[0])

Transliteration Table:

To support the transliteration of Ukrainian characters, the following table is used:

Ukrainian Transliteration
Є je
І i
Ї ji
А a
Б b
В v
Г h
Д d
Е e
Ж zh
З z
И y
Й j
К k
Л l
М m
Н n
О o
П p
Р r
С s
Т t
У u
Ф f
Х x
Ц c
Ч ch
Ш sh
Щ shch
Ь q
Ю ju
Я ja
а a
б b
в v
г h
д d
е e
ж zh
з z
и y
й j
к k
л l
м m
н n
о o
п p
р r
с s
т t
у u
ф f
х x
ц c
ч ch
ш sh
щ shch
ь q
ю ju
я ja
є je
і i
ї ji
Ґ g
ґ g
Downloads last month
53
Safetensors
Model size
144M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for tenebrisu/speecht5_tts_common_voice_uk

Finetuned
(775)
this model

Dataset used to train tenebrisu/speecht5_tts_common_voice_uk