ylacombe commited on
Commit
d369927
1 Parent(s): aac7e51

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -8
README.md CHANGED
@@ -29,18 +29,18 @@ datasets:
29
  <img src="https://huggingface.co/datasets/parler-tts/images/resolve/main/thumbnail.png" alt="Parler Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
30
 
31
 
32
- # Parler-TTS Mini v1.1 Multilingual
33
 
34
- <a target="_blank" href="https://huggingface.co/spaces/parler-tts/parler_tts">
35
  <img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-sm.svg" alt="Open in HuggingFace"/>
36
  </a>
37
 
38
- **Parler-TTS Mini v1.1 Multilingual** is a multilingual extension of [Parler-TTS Mini](https://huggingface.co/parler-tts/parler-tts-mini-v1.1).
39
 
40
  It is a fine-tuned version, trained on a [cleaned version](https://huggingface.co/datasets/PHBJT/cml-tts-cleaned-levenshtein) of [CML-TTS](https://huggingface.co/datasets/ylacombe/cml-tts) and on the non-English version of [Multilingual LibriSpeech](https://huggingface.co/datasets/facebook/multilingual_librispeech).
41
  In all, this represents some 9,200 hours of non-English data. To retain English capabilities, we also added back the [LibriTTS-R English dataset](https://huggingface.co/datasets/parler-tts/libritts_r_filtered), some 580h of high-quality English data.
42
 
43
- **Parler-TTS Mini v1.1 Multilingual** can speak in 7 European languages: English, French, Spanish, Portuguese, Polish, German, Italian and Dutch.
44
 
45
  Thanks to its **better prompt tokenizer**, it can easily be extended to other languages. This tokenizer has a larger vocabulary and handles byte fallback, which simplifies multilingual training.
46
 
@@ -79,8 +79,8 @@ import soundfile as sf
79
 
80
  device = "cuda:0" if torch.cuda.is_available() else "cpu"
81
 
82
- model = ParlerTTSForConditionalGeneration.from_pretrained("parler-tts/parler-tts-mini-v1.1").to(device)
83
- tokenizer = AutoTokenizer.from_pretrained("parler-tts/parler-tts-mini-v1.1")
84
  description_tokenizer = AutoTokenizer.from_pretrained(model.config.text_encoder._name_or_path)
85
 
86
  prompt = "Hey, how are you doing today?"
@@ -108,8 +108,8 @@ import soundfile as sf
108
 
109
  device = "cuda:0" if torch.cuda.is_available() else "cpu"
110
 
111
- model = ParlerTTSForConditionalGeneration.from_pretrained("parler-tts/parler-tts-mini-v1.1").to(device)
112
- tokenizer = AutoTokenizer.from_pretrained("parler-tts/parler-tts-mini-v1.1")
113
  description_tokenizer = AutoTokenizer.from_pretrained(model.config.text_encoder._name_or_path)
114
 
115
  prompt = "Hey, how are you doing today?"
 
29
  <img src="https://huggingface.co/datasets/parler-tts/images/resolve/main/thumbnail.png" alt="Parler Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
30
 
31
 
32
+ # Parler-TTS Mini Multilingual
33
 
34
+ <a target="_blank" href="https://huggingface.co/spaces/PHBJT/multi_parler_tts">
35
  <img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-sm.svg" alt="Open in HuggingFace"/>
36
  </a>
37
 
38
+ **Parler-TTS Mini Multilingual v1.1** is a multilingual extension of [Parler-TTS Mini](https://huggingface.co/parler-tts/parler-tts-mini-v1.1).
39
 
40
  It is a fine-tuned version, trained on a [cleaned version](https://huggingface.co/datasets/PHBJT/cml-tts-cleaned-levenshtein) of [CML-TTS](https://huggingface.co/datasets/ylacombe/cml-tts) and on the non-English version of [Multilingual LibriSpeech](https://huggingface.co/datasets/facebook/multilingual_librispeech).
41
  In all, this represents some 9,200 hours of non-English data. To retain English capabilities, we also added back the [LibriTTS-R English dataset](https://huggingface.co/datasets/parler-tts/libritts_r_filtered), some 580h of high-quality English data.
42
 
43
+ **Parler-TTS Mini Multilingual** can speak in 7 European languages: English, French, Spanish, Portuguese, Polish, German, Italian and Dutch.
44
 
45
  Thanks to its **better prompt tokenizer**, it can easily be extended to other languages. This tokenizer has a larger vocabulary and handles byte fallback, which simplifies multilingual training.
46
 
 
79
 
80
  device = "cuda:0" if torch.cuda.is_available() else "cpu"
81
 
82
+ model = ParlerTTSForConditionalGeneration.from_pretrained("parler-tts/parler-tts-mini-multilingual").to(device)
83
+ tokenizer = AutoTokenizer.from_pretrained("parler-tts/parler-tts-mini-multilingual")
84
  description_tokenizer = AutoTokenizer.from_pretrained(model.config.text_encoder._name_or_path)
85
 
86
  prompt = "Hey, how are you doing today?"
 
108
 
109
  device = "cuda:0" if torch.cuda.is_available() else "cpu"
110
 
111
+ model = ParlerTTSForConditionalGeneration.from_pretrained("parler-tts/parler-tts-mini-multilingual").to(device)
112
+ tokenizer = AutoTokenizer.from_pretrained("parler-tts/parler-tts-mini-multilingual")
113
  description_tokenizer = AutoTokenizer.from_pretrained(model.config.text_encoder._name_or_path)
114
 
115
  prompt = "Hey, how are you doing today?"