Initial commit

README.md

@@ -1,6 +1,6 @@
 ---
-title: TTS
-emoji: 🌍
+title: Coqui TTS
+emoji: 🐸
 colorFrom: indigo
 colorTo: yellow
 sdk: gradio

app.py

@@ -0,0 +1,49 @@
+import gradio as gr
+import tempfile
+from TTS.utils.synthesizer import Synthesizer
+from huggingface_hub import hf_hub_download
+
+REPO_ID = "ayymen/Coqui-TTS-Vits-shi"
+
+my_title = "Tamazight Text-to-Speech"
+my_description = "This model is based on [VITS](https://github.com/jaywalnut310/vits), thanks to 🐸 [Coqui.ai](https://coqui.ai/)." 
+
+my_examples = [
+  ["ⴰⵣⵓⵍ"],
+  ["ⴰⵔ ⴽⵎ ⵜⵜⵉⵔⵉⵖ"],
+  ["ⴳⵏ! ⴰⴷ ⴰⴽ ⵉⵙⵙⴳⵏ ⵕⴱⴱⵉ ⵉⵜⵜⵓ ⴽ."]
+]
+
+my_inputs = [
+  gr.Textbox(lines=5, label="Input Text"),
+]
+
+my_outputs = gr.Audio(type="filepath", label="Output Audio")
+
+def tts(text: str):
+    best_model_path = hf_hub_download(repo_id=REPO_ID, filename="best_model.pth") 
+    config_path = hf_hub_download(repo_id=REPO_ID, filename="config.json")
+    
+    # init synthesizer  
+    synthesizer = Synthesizer(
+        best_model_path,
+        config_path,
+        use_cuda=True
+    )
+
+    # create audio file
+    wavs = synthesizer.tts(text)
+    with tempfile.NamedTemporaryFile(suffix = ".wav", delete = False) as fp:
+        synthesizer.save_wav(wavs, fp)                      
+    return fp.name 
+ 
+iface = gr.Interface(
+    fn=tts, 
+    inputs=my_inputs, 
+    outputs=my_outputs, 
+    title=my_title, 
+    description = my_description, 
+    examples = my_examples,
+    cache_examples=True
+)
+iface.launch()

requirements.txt

@@ -0,0 +1 @@
+coqui-tts==0.24.3