Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
|
@@ -17,7 +17,8 @@ MODELS = {
|
|
| 17 |
"HKUSTAudio/Llasa-3B": "Llasa-3B",
|
| 18 |
"amphion/MaskGCT": "MaskGCT",
|
| 19 |
"OuteAI/Llama-OuteTTS-1.0-1B": "Llama-OuteTTS-1.0-1B",
|
| 20 |
-
"ByteDance/MegaTTS3": "MegaTTS3"
|
|
|
|
| 21 |
}
|
| 22 |
|
| 23 |
# Performance ratings for each model
|
|
@@ -34,6 +35,7 @@ MODEL_RATINGS = {
|
|
| 34 |
"amphion/MaskGCT": {"naturalness": "Good", "intelligibility": "Excellent", "controllability": "Moderate"},
|
| 35 |
"OuteAI/Llama-OuteTTS-1.0-1B": {"naturalness": "Moderate", "intelligibility": "Moderate", "controllability": "Moderate"},
|
| 36 |
"ByteDance/MegaTTS3": {"naturalness": "Good", "intelligibility": "Good", "controllability": "Moderate"}
|
|
|
|
| 37 |
}
|
| 38 |
|
| 39 |
# Model descriptions for better understanding
|
|
@@ -50,6 +52,7 @@ MODEL_DESCRIPTIONS = {
|
|
| 50 |
"amphion/MaskGCT": "Masked generative modeling approach",
|
| 51 |
"OuteAI/Llama-OuteTTS-1.0-1B": "LLM-based TTS with moderate performance",
|
| 52 |
"ByteDance/MegaTTS3": "Industrial-grade TTS solution"
|
|
|
|
| 53 |
}
|
| 54 |
|
| 55 |
# Folder that contains subfolders with the audio clips
|
|
|
|
| 17 |
"HKUSTAudio/Llasa-3B": "Llasa-3B",
|
| 18 |
"amphion/MaskGCT": "MaskGCT",
|
| 19 |
"OuteAI/Llama-OuteTTS-1.0-1B": "Llama-OuteTTS-1.0-1B",
|
| 20 |
+
"ByteDance/MegaTTS3": "MegaTTS3",
|
| 21 |
+
"Kyutai/Kyutai-TTS"
|
| 22 |
}
|
| 23 |
|
| 24 |
# Performance ratings for each model
|
|
|
|
| 35 |
"amphion/MaskGCT": {"naturalness": "Good", "intelligibility": "Excellent", "controllability": "Moderate"},
|
| 36 |
"OuteAI/Llama-OuteTTS-1.0-1B": {"naturalness": "Moderate", "intelligibility": "Moderate", "controllability": "Moderate"},
|
| 37 |
"ByteDance/MegaTTS3": {"naturalness": "Good", "intelligibility": "Good", "controllability": "Moderate"}
|
| 38 |
+
"Kyutai/Kyutai-TTS": {"naturalness": "Good", "intelligibility": "Good", "controllability": "Moderate"}
|
| 39 |
}
|
| 40 |
|
| 41 |
# Model descriptions for better understanding
|
|
|
|
| 52 |
"amphion/MaskGCT": "Masked generative modeling approach",
|
| 53 |
"OuteAI/Llama-OuteTTS-1.0-1B": "LLM-based TTS with moderate performance",
|
| 54 |
"ByteDance/MegaTTS3": "Industrial-grade TTS solution"
|
| 55 |
+
"Kyutai/Kyutai-TTS": "Industrial-grade TTS solution"
|
| 56 |
}
|
| 57 |
|
| 58 |
# Folder that contains subfolders with the audio clips
|