Upload VibeVoice French fine-tuned model (SIWIS, 10 epochs, full FT)

README.md

@@ -0,0 +1,74 @@
+---
+license: mit
+base_model: microsoft/VibeVoice-Realtime-0.5B
+tags:
+  - tts
+  - text-to-speech
+  - french
+  - vibevoice
+  - finetuned
+language:
+  - fr
+datasets:
+  - custom
+pipeline_tag: text-to-speech
+---
+
+# VibeVoice-Realtime-0.5B Fine-tuned (French SIWIS)
+
+Fine-tuned version of [microsoft/VibeVoice-Realtime-0.5B](https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B) on the French SIWIS dataset for improved French TTS.
+
+## Training Details
+
+- **Base model**: microsoft/VibeVoice-Realtime-0.5B
+- **Training data**: SIWIS French Speech Synthesis Database (~9,200 samples, 500 benchmark phrases excluded)
+- **Training type**: Full fine-tuning of TTS language model (434M params)
+- **Frozen components**: Acoustic tokenizer (VAE), prediction head (diffusion), language encoder (Qwen2.5 4 layers)
+
+### Hyperparameters
+
+| Parameter | Value |
+|-----------|-------|
+| Epochs | 10 |
+| Batch size | 4 |
+| Gradient accumulation | 4 |
+| Effective batch size | 16 |
+| Learning rate | 5e-5 |
+| Weight decay | 0.01 |
+| Warmup steps | 500 |
+| Precision | bf16 |
+
+### Hardware
+
+- GPU: NVIDIA RTX 6000 Ada (49GB)
+
+## Benchmark Results (500 SIWIS French phrases)
+
+| Metric | Value |
+|--------|-------|
+| WER (mean) | 35.0% |
+| WER (median) | 22.9% |
+| RTF (mean) | 0.416 |
+
+## Usage
+
+```python
+import torch
+import soundfile as sf
+from vibevoice.modular.modeling_vibevoice_streaming_inference import (
+    VibeVoiceStreamingForConditionalGenerationInference,
+)
+
+model = VibeVoiceStreamingForConditionalGenerationInference.from_pretrained(
+    "Rcarvalo/vibevoice",
+    torch_dtype=torch.bfloat16,
+).to("cuda")
+
+# Generate French speech
+audio = model.generate(text="Bonjour, comment allez-vous aujourd'hui?")
+sf.write("output.wav", audio.cpu().numpy(), 24000)
+```
+
+## License
+
+MIT (same as base model)

config.json

@@ -0,0 +1,86 @@
+{
+  "acoustic_vae_dim": 64,
+  "acoustic_tokenizer_config": {
+    "causal": true,
+    "channels": 1,
+    "conv_bias": true,
+    "conv_norm": "none",
+    "corpus_normalize": 0.0,
+    "decoder_depths": null,
+    "decoder_n_filters": 32,
+    "decoder_ratios": [
+      8,
+      5,
+      5,
+      4,
+      2,
+      2
+    ],
+    "disable_last_norm": true,
+    "encoder_depths": "3-3-3-3-3-3-8",
+    "encoder_n_filters": 32,
+    "encoder_ratios": [
+      8,
+      5,
+      5,
+      4,
+      2,
+      2
+    ],
+    "fix_std": 0.5,
+    "layer_scale_init_value": 1e-06,
+    "layernorm": "RMSNorm",
+    "layernorm_elementwise_affine": true,
+    "layernorm_eps": 1e-05,
+    "mixer_layer": "depthwise_conv",
+    "model_type": "vibevoice_acoustic_tokenizer",
+    "pad_mode": "constant",
+    "std_dist_type": "gaussian",
+    "vae_dim": 64,
+    "weight_init_value": 0.01
+  },
+  "architectures": [
+    "VibeVoiceStreamingForConditionalGenerationInference"
+  ],
+  "decoder_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "silu",
+    "hidden_size": 896,
+    "initializer_range": 0.02,
+    "intermediate_size": 4864,
+    "max_position_embeddings": 8192,
+    "max_window_layers": 24,
+    "model_type": "qwen2",
+    "num_attention_heads": 14,
+    "num_hidden_layers": 24,
+    "num_key_value_heads": 2,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000.0,
+    "sliding_window": null,
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "use_cache": true,
+    "use_sliding_window": false,
+    "vocab_size": 151936
+  },
+  "diffusion_head_config": {
+    "ddpm_batch_mul": 4,
+    "ddpm_beta_schedule": "cosine",
+    "ddpm_num_inference_steps": 20,
+    "ddpm_num_steps": 1000,
+    "diffusion_type": "ddpm",
+    "head_ffn_ratio": 3.0,
+    "head_layers": 4,
+    "hidden_size": 896,
+    "latent_size": 64,
+    "model_type": "vibevoice_diffusion_head",
+    "prediction_type": "v_prediction",
+    "rms_norm_eps": 1e-05,
+    "speech_vae_dim": 64
+  },
+  "model_type": "vibevoice_streaming",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "tts_backbone_num_hidden_layers": 20
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dbfa06e998ae8f5130475768bd5dafe56b979e420f289ca80807df16158240d
+size 1737558440

preprocessor_config.json

@@ -0,0 +1,13 @@
+{
+  "processor_class": "VibeVoiceStreamingProcessor",
+  "speech_tok_compress_ratio": 3200,
+  "db_normalize": true,
+  "audio_processor": {
+    "feature_extractor_type": "VibeVoiceTokenizerProcessor",
+    "sampling_rate": 24000,
+    "normalize_audio": true,
+    "target_dB_FS": -25,
+    "eps": 1e-06
+  },
+  "language_model_pretrained_name": "Qwen/Qwen2.5-0.5B"
+}

train_config.json

@@ -0,0 +1,20 @@
+{
+  "data_dir": "data/tokenized_siwis,data/tokenized_siwis_full",
+  "val_dir": "data/tokenized_val_full",
+  "output_dir": "outputs/full_ft_vibevoice",
+  "epochs": 10,
+  "batch_size": 4,
+  "grad_accum": 4,
+  "lr": 5e-05,
+  "weight_decay": 0.01,
+  "warmup_steps": 500,
+  "full_finetune": true,
+  "lora_rank": 16,
+  "lora_alpha": 32.0,
+  "max_text_len": 256,
+  "max_speech_len": 120,
+  "device": "cuda",
+  "bf16": true,
+  "save_every": 500,
+  "eval_every": 500
+}