Merge pull request #59 from ks6088ts-labs/feature/issue-58_tts-app

add text to speech app

.env.template

@@ -4,6 +4,8 @@ AZURE_OPENAI_API_KEY="<YOUR_API_KEY>"
 AZURE_OPENAI_API_VERSION="2024-06-01"
 AZURE_OPENAI_GPT_MODEL="gpt-4o"
 AZURE_OPENAI_STT_MODEL="whisper"
+AZURE_OPENAI_TTS_VOICE="alloy"
+AZURE_OPENAI_TTS_MODEL="tts-hd"
 AZURE_OPENAI_EMBEDDING_MODEL="text-embedding-3-large"
 AZURE_OPENAI_DALLE_MODEL="dall-e-3"
 

.gitignore

@@ -162,3 +162,4 @@ cython_debug/
 # Project
 *.env
 generated/
+*.mp3

apps/99_streamlit_examples/README.md

@@ -57,11 +57,15 @@ Access to http://localhost:8501 and select the sample you want to run from the s
 
 ![Speech to Text](../../docs/images/99_streamlit_examples.stt.png)
 
-#### 7. Create image
+#### 7. Text to Speech
+
+![Text to Speech](../../docs/images/99_streamlit_examples.tts.png)
+
+#### 8. Create image
 
 ![Create image](../../docs/images/99_streamlit_examples.createimage.png)
 
-#### 8. Visualize location
+#### 9. Visualize location
 
 ![Visualize location](../../docs/images/99_streamlit_examples.map.png)
 

apps/99_streamlit_examples/pages/7_Text_to_speech.py

@@ -0,0 +1,72 @@
+from os import getenv
+
+import streamlit as st
+from dotenv import load_dotenv
+from openai import AzureOpenAI
+
+load_dotenv()
+
+with st.sidebar:
+    azure_openai_endpoint = st.text_input(
+        label="AZURE_OPENAI_ENDPOINT",
+        value=getenv("AZURE_OPENAI_ENDPOINT"),
+        key="AZURE_OPENAI_ENDPOINT",
+        type="default",
+    )
+    azure_openai_api_key = st.text_input(
+        label="AZURE_OPENAI_API_KEY",
+        key="AZURE_OPENAI_API_KEY",
+        type="password",
+    )
+    azure_openai_api_version = st.text_input(
+        label="AZURE_OPENAI_API_VERSION",
+        value=getenv("AZURE_OPENAI_API_VERSION"),
+        key="AZURE_OPENAI_API_VERSION",
+        type="default",
+    )
+    azure_openai_tts_model = st.text_input(
+        label="AZURE_OPENAI_TTS_MODEL",
+        value=getenv("AZURE_OPENAI_TTS_MODEL"),
+        key="AZURE_OPENAI_TTS_MODEL",
+        type="default",
+    )
+    "[Azure Portal](https://portal.azure.com/)"
+    "[Azure OpenAI Studio](https://oai.azure.com/resource/overview)"
+    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/99_streamlit_examples/pages/7_Text_to_speech.py)"
+
+st.title("Text to speech")
+
+if not azure_openai_api_key or not azure_openai_endpoint or not azure_openai_api_version or not azure_openai_tts_model:
+    st.warning("Please fill in the required fields at the sidebar.")
+    st.stop()
+
+st.info("This is a sample to convert text to speech.")
+
+content = st.text_area(
+    label="Text to speech",
+    value="",
+)
+st.write(f"You wrote {len(content)} characters.")
+
+if st.button("Convert"):
+    client = AzureOpenAI(
+        api_key=azure_openai_api_key,
+        api_version=azure_openai_api_version,
+        azure_endpoint=azure_openai_endpoint,
+    )
+    path_to_output = "output.mp3"
+
+    with st.spinner("Converting..."):
+        response = client.audio.speech.create(
+            input=content,
+            voice=getenv("AZURE_OPENAI_TTS_VOICE"),
+            model=azure_openai_tts_model,
+            response_format="mp3",
+        )
+        response.write_to_file(path_to_output)
+
+    st.audio(
+        data=path_to_output,
+        format="audio/mpeg",
+        loop=False,
+    )

apps/99_streamlit_examples/pages/7_Create_image.py → apps/99_streamlit_examples/pages/8_Create_image.py

@@ -33,7 +33,7 @@
     )
     "[Azure Portal](https://portal.azure.com/)"
     "[Azure OpenAI Studio](https://oai.azure.com/resource/overview)"
-    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/99_streamlit_examples/pages/7_Create_image.py)"
+    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/99_streamlit_examples/pages/8_Create_image.py)"
 
 st.title("Create image")
 

apps/99_streamlit_examples/pages/8_Visualize_location.py → apps/99_streamlit_examples/pages/9_Visualize_location.py

@@ -8,7 +8,7 @@
 with st.sidebar:
     "[Azure Portal](https://portal.azure.com/)"
     "[Azure OpenAI Studio](https://oai.azure.com/resource/overview)"
-    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/99_streamlit_examples/pages/8_Visualize_location.py)"
+    "[View the source code](https://github.com/ks6088ts-labs/workshop-azure-openai/blob/main/apps/99_streamlit_examples/pages/9_Visualize_location.py)"
 
 st.title("Visualize location")
 st.info("This is a sample to visualize location.")

tests/test_smoke.py

@@ -9,6 +9,15 @@ def test_smoke():
         # "apps/7_streamlit_chat_rag/main.py",
         "apps/8_streamlit_azure_openai_batch/main.py",
         "apps/99_streamlit_examples/main.py",
+        "apps/99_streamlit_examples/pages/1_File_Q&A.py",
+        "apps/99_streamlit_examples/pages/2_Image_Q&A.py",
+        "apps/99_streamlit_examples/pages/3_Camera_Q&A.py",
+        "apps/99_streamlit_examples/pages/4_Translate_text.py",
+        "apps/99_streamlit_examples/pages/5_Explain_data.py",
+        "apps/99_streamlit_examples/pages/6_Speech_to_text.py",
+        "apps/99_streamlit_examples/pages/7_Text_to_speech.py",
+        "apps/99_streamlit_examples/pages/8_Create_image.py",
+        "apps/99_streamlit_examples/pages/9_Visualize_location.py",
     ]
     for path in paths:
         at = AppTest(