Update README.md
Browse files
README.md
CHANGED
@@ -184,18 +184,19 @@ import soundfile as sf
|
|
184 |
|
185 |
device = "cuda:0" if torch.cuda.is_available() else "cpu"
|
186 |
|
187 |
-
model = ParlerTTSForConditionalGeneration.from_pretrained("
|
188 |
-
tokenizer = AutoTokenizer.from_pretrained("
|
|
|
189 |
|
190 |
prompt = "अरे, तुम आज कैसे हो?"
|
191 |
description = "Divya's voice is monotone yet slightly fast in delivery, with a very close recording that almost has no background noise."
|
192 |
|
193 |
-
input_ids =
|
194 |
prompt_input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(device)
|
195 |
|
196 |
generation = model.generate(input_ids=input_ids, prompt_input_ids=prompt_input_ids)
|
197 |
audio_arr = generation.cpu().numpy().squeeze()
|
198 |
-
sf.write("
|
199 |
```
|
200 |
|
201 |
The model includes **69 speakers** across 18 officially supported languages, with each language having a set of recommended voices for optimal performance. Below is a table summarizing the available speakers for each language, along with the recommended ones.
|
|
|
184 |
|
185 |
device = "cuda:0" if torch.cuda.is_available() else "cpu"
|
186 |
|
187 |
+
model = ParlerTTSForConditionalGeneration.from_pretrained("ai4bharat/indic-parler-tts").to(device)
|
188 |
+
tokenizer = AutoTokenizer.from_pretrained("ai4bharat/indic-parler-tts")
|
189 |
+
description_tokenizer = AutoTokenizer.from_pretrained(model.config.text_encoder._name_or_path)
|
190 |
|
191 |
prompt = "अरे, तुम आज कैसे हो?"
|
192 |
description = "Divya's voice is monotone yet slightly fast in delivery, with a very close recording that almost has no background noise."
|
193 |
|
194 |
+
input_ids = description_tokenizer(description, return_tensors="pt").input_ids.to(device)
|
195 |
prompt_input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to(device)
|
196 |
|
197 |
generation = model.generate(input_ids=input_ids, prompt_input_ids=prompt_input_ids)
|
198 |
audio_arr = generation.cpu().numpy().squeeze()
|
199 |
+
sf.write("indic_tts_out.wav", audio_arr, model.config.sampling_rate)
|
200 |
```
|
201 |
|
202 |
The model includes **69 speakers** across 18 officially supported languages, with each language having a set of recommended voices for optimal performance. Below is a table summarizing the available speakers for each language, along with the recommended ones.
|