progress

firebase · Aug 7, 2024 · 8e6dafc · 8e6dafc
1 parent eec9636
commit 8e6dafc
Show file tree

Hide file tree

Showing 7 changed files with 229 additions and 189 deletions.
diff --git a/vertexai/app/build.gradle.kts b/vertexai/app/build.gradle.kts
@@ -57,6 +57,7 @@ dependencies {
     implementation("androidx.activity:activity-compose:1.9.1")
     implementation("androidx.navigation:navigation-compose:2.7.7")
 
+    implementation("androidx.compose.runtime:runtime-livedata:1.3.0")
     implementation(platform("androidx.compose:compose-bom:2024.06.00"))
     implementation("androidx.compose.ui:ui")
     implementation("androidx.compose.ui:ui-graphics")

diff --git a/vertexai/app/src/main/AndroidManifest.xml b/vertexai/app/src/main/AndroidManifest.xml
@@ -16,6 +16,8 @@
 <manifest xmlns:android="http://schemas.android.com/apk/res/android"
     xmlns:tools="http://schemas.android.com/tools">
 
+    <uses-permission android:name="android.permission.RECORD_AUDIO"/>
+
     <application
         android:allowBackup="true"
         android:dataExtractionRules="@xml/data_extraction_rules"

diff --git a/...p/src/main/kotlin/com/google/firebase/quickstart/vertexai/GenerativeAiViewModelFactory.kt b/...p/src/main/kotlin/com/google/firebase/quickstart/vertexai/GenerativeAiViewModelFactory.kt
@@ -56,7 +56,7 @@ val GenerativeViewModelFactory = object : ViewModelProvider.Factory {
                     // Initialize a GenerativeModel with the `gemini-flash` AI model
                     // for multimodal text generation
                     val generativeModel = Firebase.vertexAI.generativeModel(
-                        modelName = "gemini-1.5-flash-preview-0514",
+                        modelName = "gemini-1.5-pro-001",
                         generationConfig = config
                     )
                     PhotoReasoningViewModel(generativeModel)

diff --git a/...c/main/kotlin/com/google/firebase/quickstart/vertexai/feature/multimodal/AudioRecorder.kt b/...c/main/kotlin/com/google/firebase/quickstart/vertexai/feature/multimodal/AudioRecorder.kt
@@ -0,0 +1,89 @@
+package com.google.firebase.quickstart.vertexai.feature.multimodal
+
+import android.content.ContentValues
+import android.content.Context
+import android.media.MediaRecorder
+import android.os.Build
+import android.os.Environment
+import android.provider.MediaStore
+import androidx.core.app.ActivityCompat
+import android.Manifest
+import android.content.pm.PackageManager
+import java.io.File
+import java.io.IOException
+
+class AudioRecorder(private val context: Context) {
+  private var recorder: MediaRecorder? = null
+  private var outputFilePath: String? = null
+
+  fun startRecording() {
+    if (ActivityCompat.checkSelfPermission(context, Manifest.permission.RECORD_AUDIO) != PackageManager.PERMISSION_GRANTED) {
+      println("Manifest.permission.RECORD_AUDIO) != PackageManager.PERMISSION_GRANTED")
+    }
+    val audioFileName = "recording_${System.currentTimeMillis()}.m4a"
+    if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.S) {
+      val values = ContentValues().apply {
+        put(MediaStore.Audio.Media.RELATIVE_PATH, Environment.DIRECTORY_MUSIC)
+        put(MediaStore.Audio.Media.TITLE, audioFileName)
+        put(MediaStore.Audio.Media.MIME_TYPE, "audio/mp4")
+        put(MediaStore.Audio.Media.IS_PENDING, 1)
+      }
+
+      val audioCollection = MediaStore.Audio.Media.getContentUri(MediaStore.VOLUME_EXTERNAL_PRIMARY)
+      val audioUri = context.contentResolver.insert(audioCollection, values)
+      audioUri?.let { uri ->
+        outputFilePath = uri.toString()
+        context.contentResolver.openFileDescriptor(uri, "w")?.use { pfd ->
+          recorder = MediaRecorder(context).apply {
+            setAudioSource(MediaRecorder.AudioSource.MIC)
+            setOutputFormat(MediaRecorder.OutputFormat.MPEG_4)
+            setAudioEncoder(MediaRecorder.AudioEncoder.AAC)
+            setOutputFile(pfd.fileDescriptor)
+            try {
+              prepare()
+            } catch (e: IOException) {
+              e.printStackTrace()
+            }
+            start()
+          }
+        }
+      }
+    } else {
+      outputFilePath = "${context.getExternalFilesDir(Environment.DIRECTORY_MUSIC)}/$audioFileName"
+      recorder = MediaRecorder().apply {
+        setAudioSource(MediaRecorder.AudioSource.MIC)
+        setOutputFormat(MediaRecorder.OutputFormat.MPEG_4)
+        setAudioEncoder(MediaRecorder.AudioEncoder.AAC)
+        setOutputFile(outputFilePath)
+        try {
+          prepare()
+        } catch (e: IOException) {
+          e.printStackTrace()
+        }
+        start()
+      }
+    }
+  }
+
+  fun stopRecording(): ByteArray? {
+    recorder?.apply {
+      stop()
+      release()
+    }
+    recorder = null
+
+    outputFilePath?.let { path ->
+      return if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.Q) {
+        val uri = android.net.Uri.parse(path)
+        val values = ContentValues().apply {
+          put(MediaStore.Audio.Media.IS_PENDING, 0)
+        }
+        context.contentResolver.update(uri, values, null, null)
+        context.contentResolver.openInputStream(uri)?.readBytes()
+      } else {
+        File(path).readBytes()
+      }
+    }
+    return null
+  }
+}