(wip)debug

templates/arena.html

@@ -26,6 +26,7 @@
                 <input type="file" id="voice-file" accept="audio/*">
                 <audio id="voice-preview" controls style="display:none;"></audio>
             </div>
+            <hr>
             <div class="input-group">
                 <button type="button" class="segmented-btn random-btn" title="Roll random text">
                     <svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" viewBox="0 0 24 24" fill="none"
@@ -45,7 +46,8 @@
         </form>
 
         <div id="initial-keyboard-hint" class="keyboard-hint">
-            Press <kbd>R</kbd> for random text, <kbd>N</kbd> for next random round, <kbd>Enter</kbd> to generate
+            Press <kbd>R</kbd> for random text, <kbd>V</kbd> for random reference voice, <kbd>N</kbd> for next random
+            round, <kbd>Enter</kbd> to generate
         </div>
 
         <div class="loading-container" style="display: none;">
@@ -107,7 +109,8 @@
             <button class="next-round-btn">Next Round</button>
         </div>
         <div id="playback-keyboard-hint" class="keyboard-hint" style="display: none;">
-            Press <kbd>Space</kbd> to play/pause, <kbd>A</kbd>/<kbd>B</kbd> to vote, <kbd>R</kbd> for random text, <kbd>N</kbd>
+            Press <kbd>Space</kbd> to play/pause, <kbd>A</kbd>/<kbd>B</kbd> to vote, <kbd>R</kbd> for random text, <kbd>V</kbd>
+            for random reference voice, <kbd>N</kbd>
             for next random round
         </div>
     </div>
@@ -141,18 +144,13 @@
 
         .random-voice-btn {
             height: 36px;
-            width: 36px;
-            background-color: white;
+            width: 48px;
             border: 1px solid var(--border-color);
             border-radius: var(--radius);
             margin-right: 10px;
             flex-shrink: 0;
         }
 
-        .random-voice-btn:hover {
-            background-color: var(--light-gray);
-        }
-
         .random-voice-btn svg {
             color: var(--primary-color);
         }
@@ -629,51 +627,7 @@
                 const voiceFileInput = document.getElementById('voice-file');
                 const voicePreview = document.getElementById('voice-preview');
                 if (randomVoiceBtn && voiceFileInput && voicePreview) {
-                    randomVoiceBtn.addEventListener('click', function () {
-                        // 显示加载状态
-                        randomVoiceBtn.classList.add('loading');
-
-                        // 获取随机参考音色
-                        fetch('/api/voice/random')
-                            .then(response => {
-                                if (!response.ok) {
-                                    throw new Error('获取随机音色失败');
-                                }
-                                return response.blob();
-                            })
-                            .then(audioBlob => {
-                                // 创建文件对象，用于合成时提交
-                                const fileName = 'random_voice_sample.' +
-                                    (audioBlob.type.split('/')[1] || 'mp3');
-
-                                // 创建File对象，用于后续上传
-                                const audioFile = new File([audioBlob], fileName, {type: audioBlob.type});
-
-                                // 创建一个DataTransfer对象来模拟文件输入
-                                const dataTransfer = new DataTransfer();
-                                dataTransfer.items.add(audioFile);
-                                voiceFileInput.files = dataTransfer.files;
-
-                                // 更新音频预览
-                                const audioUrl = URL.createObjectURL(audioBlob);
-                                voicePreview.src = audioUrl;
-                                voicePreview.style.display = 'inline-block';
-                                voicePreview.load();
-                                voicePreview.play();
-
-                                // 触发change事件，确保其他监听器知道文件已更改
-                                const event = new Event('change', {bubbles: true});
-                                voiceFileInput.dispatchEvent(event);
-                            })
-                            .catch(error => {
-                                console.error('获取随机音色出错:', error);
-                                openToast ? openToast("获取随机参考音色失败", "error") : alert("获取随机参考音色失败");
-                            })
-                            .finally(() => {
-                                // 移除加载状态
-                                randomVoiceBtn.classList.remove('loading');
-                            });
-                    });
+                    randomVoiceBtn.addEventListener('click', handleRandomVoice);
                     voiceFileInput.addEventListener('change', function () {
                         const file = this.files[0];
                         if (file) {
@@ -1056,6 +1010,56 @@
                     textInput.focus();
                 }
 
+                function handleRandomVoice() {
+                    const randomVoiceBtn = document.querySelector('.random-voice-btn');
+                    const voiceFileInput = document.getElementById('voice-file');
+                    const voicePreview = document.getElementById('voice-preview');
+
+                    // 显示加载状态
+                    randomVoiceBtn.classList.add('loading');
+
+                    // 获取随机参考音色
+                    fetch('/api/voice/random')
+                        .then(response => {
+                            if (!response.ok) {
+                                throw new Error('获取随机音色失败');
+                            }
+                            return response.blob();
+                        })
+                        .then(audioBlob => {
+                            // 创建文件对象，用于合成时提交
+                            const fileName = 'random_voice_sample.' +
+                                (audioBlob.type.split('/')[1] || 'mp3');
+
+                            // 创建File对象，用于后续上传
+                            const audioFile = new File([audioBlob], fileName, {type: audioBlob.type});
+
+                            // 创建一个DataTransfer对象来模拟文件输入
+                            const dataTransfer = new DataTransfer();
+                            dataTransfer.items.add(audioFile);
+                            voiceFileInput.files = dataTransfer.files;
+
+                            // 更新音频预览
+                            const audioUrl = URL.createObjectURL(audioBlob);
+                            voicePreview.src = audioUrl;
+                            voicePreview.style.display = 'inline-block';
+                            voicePreview.load();
+                            voicePreview.play();
+
+                            // 触发change事件，确保其他监听器知道文件已更改
+                            const event = new Event('change', {bubbles: true});
+                            voiceFileInput.dispatchEvent(event);
+                        })
+                        .catch(error => {
+                            console.error('获取随机音色出错:', error);
+                            openToast ? openToast("获取随机参考音色失败", "error") : alert("获取随机参考音色失败");
+                        })
+                        .finally(() => {
+                            // 移除加载状态
+                            randomVoiceBtn.classList.remove('loading');
+                        });
+                }
+
                 function showListenToastMessage() {
                     openToast("Please listen to both audio samples before voting", "info");
                 }
@@ -1139,6 +1143,11 @@
                             e.preventDefault();
                             handleRandom();
                         }
+                    } else if (e.key.toLowerCase() === 'v') {
+                        if (!e.ctrlKey && !e.metaKey && !e.altKey) {
+                            e.preventDefault();
+                            handleRandomVoice();
+                        }
                     } else if (e.key === ' ') {
                         // Space to play/pause current audio
                         if (playersContainer.style.display !== 'none') {

tts.py

@@ -47,7 +47,7 @@ data = {"text": "string", "provider": "string", "model": "string"}
 
 def predict_index_tts(text, reference_audio_path=None):
     from gradio_client import Client, handle_file
-    client = Client("kemuriririn/IndexTTS")
+    client = Client("kemuriririn/IndexTTS",verbose=True)
     if reference_audio_path:
         prompt = handle_file(reference_audio_path)
     else: