Add code

app.py

@@ -14,12 +14,21 @@ from pydub import AudioSegment
 from transformers import AutoTokenizer, AutoFeatureExtractor, set_seed
 from transformers.generation.streamers import BaseStreamer
 from huggingface_hub import InferenceClient
+import nltk
+nltk.download('punkt')
+
 
 device = "cuda:0" if torch.cuda.is_available() else "mps" if torch.backends.mps.is_available() else "cpu"
 torch_dtype = torch.float16 if device != "cpu" else torch.float32
 
 repo_id = "parler-tts/parler_tts_mini_v0.1"
 
+jenny_repo_id = "ylacombe/parler-tts-mini-jenny-30H"
+
+model = ParlerTTSForConditionalGeneration.from_pretrained(
+    jenny_repo_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
+).to(device)
+
 model = ParlerTTSForConditionalGeneration.from_pretrained(
     repo_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True
 ).to(device)
@@ -205,6 +214,9 @@ def generate_base(subject, setting, ):
     gr.Info("Story Generated", duration=3)
     story = response.choices[0].message.content
 
+    model_input = story.replace("\n", " ").strip()
+    model_input = nltk.sent_tokenize(model_input)
+
     play_steps_in_s = 4.0
     play_steps = int(frame_rate * play_steps_in_s)
     streamer = ParlerTTSStreamer(model, device=device, play_steps=play_steps)

requirements.txt

@@ -1,3 +1,4 @@
 https://gradio-builds.s3.amazonaws.com/bed454c3d22cfacedc047eb3b0ba987b485ac3fd/gradio-4.40.0-py3-none-any.whl
 git+https://github.com/huggingface/parler-tts.git
-accelerate
+accelerate
+nltk