scyllasband / litert /manifest.json
spybyscript's picture
Add optional INT8 ONNX bundle
49abc9a verified
Raw
History Blame Contribute Delete
21.6 kB
{
"architecture": "scyllasband-duration-flow",
"assets": {
"emotion_index": "assets/emotions.json",
"g2p_config": "assets/g2p/export_config.json",
"g2p_language_map": "assets/g2p/language_map.json",
"g2p_normalization": "assets/g2p/normalization.json",
"g2p_phoneme_dict": "assets/g2p/phoneme_dict.json",
"g2p_pronunciation_overrides": "assets/g2p/pronunciation_overrides.json",
"g2p_tokenizer": "assets/g2p/tokenizer.json",
"language_index": "assets/languages.json",
"phone_vocab": "assets/phone_vocab.json",
"voice_index": "assets/voices.json",
"voice_pack_manifest": "assets/voice_packs/reference_packs.json",
"voice_packs": "assets/voice_packs"
},
"audio": {
"hop_length": 256,
"latent_dim": 24,
"latent_hop_length": 512,
"n_mels": 100,
"sample_rate": 24000
},
"components": {
"duration_predictor": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/duration_predictor.tflite"
}
},
"inputs": [
"phone_ids",
"voice_id",
"language_id",
"affect_values",
"boundary_before_id",
"boundary_after_id",
"phone_mask",
"affect_condition_mask",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"durations"
],
"required": true
},
"g2p": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/g2p.tflite"
}
},
"inputs": [
"text",
"language"
],
"outputs": [
"phones"
],
"required": true
},
"vector_context_encoder": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vector_context_encoder.tflite"
}
},
"inputs": [
"span_context_phone_ids",
"span_context_segment_ids",
"span_context_mask"
],
"outputs": [
"span_context_hidden"
],
"required": true
},
"vector_estimator": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vector_estimator.tflite"
}
},
"inputs": [
"noise",
"time",
"expanded_phone_ids",
"voice_id",
"language_id",
"affect_values",
"boundary_before_id",
"boundary_after_id",
"latent_mask",
"affect_condition_mask",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask",
"prefix_latents",
"prefix_mask",
"span_context_hidden"
],
"outputs": [
"latent_velocity"
],
"required": true
},
"vector_estimator_256": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vector_estimator_256.tflite"
}
},
"inputs": [
"noise",
"time",
"expanded_phone_ids",
"voice_id",
"language_id",
"affect_values",
"boundary_before_id",
"boundary_after_id",
"latent_mask",
"affect_condition_mask",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask",
"prefix_latents",
"prefix_mask",
"span_context_hidden"
],
"outputs": [
"velocity"
],
"required": false
},
"vector_estimator_384": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vector_estimator_384.tflite"
}
},
"inputs": [
"noise",
"time",
"expanded_phone_ids",
"voice_id",
"language_id",
"affect_values",
"boundary_before_id",
"boundary_after_id",
"latent_mask",
"affect_condition_mask",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask",
"prefix_latents",
"prefix_mask",
"span_context_hidden"
],
"outputs": [
"velocity"
],
"required": false
},
"vector_estimator_512": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vector_estimator_512.tflite"
}
},
"inputs": [
"noise",
"time",
"expanded_phone_ids",
"voice_id",
"language_id",
"affect_values",
"boundary_before_id",
"boundary_after_id",
"latent_mask",
"affect_condition_mask",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask",
"prefix_latents",
"prefix_mask",
"span_context_hidden"
],
"outputs": [
"velocity"
],
"required": false
},
"vocoder": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"waveform"
],
"required": true
},
"vocoder_256": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_256.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"audio"
],
"required": false
},
"vocoder_384": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_384.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"audio"
],
"required": false
},
"vocoder_512": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_512.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"audio"
],
"required": false
},
"vocoder_adapter": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_adapter.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"vocos_features"
],
"required": false
},
"vocoder_adapter_256": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_adapter_256.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"vocos_features"
],
"required": false
},
"vocoder_adapter_384": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_adapter_384.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"vocos_features"
],
"required": false
},
"vocoder_adapter_512": {
"artifacts": {
"litert": {
"format": "litert",
"path": "litert/vocoder_adapter_512.tflite"
}
},
"inputs": [
"latents",
"latent_mask",
"voice_id",
"language_id",
"emotion_id",
"reference_style",
"reference_prosody",
"reference_mask",
"reference_condition_mask"
],
"outputs": [
"vocos_features"
],
"required": false
}
},
"contract_version": "1.0.0",
"controls": {
"affect": {
"all_zero_is_explicit": true,
"axes": [
"calm",
"joy",
"anger",
"sadness",
"sarcasm",
"questioning"
],
"axis_order_version": 1,
"condition_mask_input": true,
"default_preset": "calm",
"dimension": 6,
"enabled": true,
"guidance": {
"default_scale": 1.0,
"enabled": true,
"host_linear_combination": true,
"identity_reference_retained": true,
"minimum_scale": 0.0,
"null_condition_mask": 0.0
},
"legacy_presets": {
"angry": [
0.0,
0.0,
0.75,
0.0,
0.0,
0.0
],
"excited": [
0.0,
1.0,
0.25,
0.0,
0.0,
0.0
],
"friendly": [
0.5,
0.5,
0.0,
0.0,
0.0,
0.0
],
"frustrated": [
0.0,
0.0,
0.5,
0.25,
0.0,
0.0
],
"happy": [
0.0,
0.75,
0.0,
0.0,
0.0,
0.0
],
"neutral": [
0.5,
0.0,
0.0,
0.0,
0.0,
0.0
],
"sad": [
0.0,
0.0,
0.0,
0.75,
0.0,
0.0
],
"sarcastic": [
0.25,
0.0,
0.0,
0.0,
0.75,
0.0
],
"unsure": [
0.25,
0.0,
0.0,
0.0,
0.0,
0.75
]
},
"preset_version": 2,
"presets": {
"anger": [
0.0,
0.0,
0.75,
0.0,
0.0,
0.0
],
"calm": [
0.5,
0.0,
0.0,
0.0,
0.0,
0.0
],
"joy": [
0.0,
0.75,
0.0,
0.0,
0.0,
0.0
],
"questioning": [
0.25,
0.0,
0.0,
0.0,
0.0,
0.75
],
"sadness": [
0.0,
0.0,
0.0,
0.75,
0.0,
0.0
],
"sarcasm": [
0.25,
0.0,
0.0,
0.0,
0.75,
0.0
]
},
"range": [
0.0,
1.0
]
},
"chunk_context": {
"artificial_continuation_boundaries": [
"chunk_continue"
],
"artificial_continuation_phone": null,
"boundary_after_values": [
"sentence_end",
"paragraph_end",
"clause_continue",
"chunk_continue"
],
"boundary_before_values": [
"sentence_start",
"paragraph_start",
"clause_continue",
"chunk_continue"
],
"clause_continuation_fallback_phone": "<pause_comma>",
"conditioning": "boundary_embeddings",
"context_phone_tokens": {
"continuation": "<ctx_continuation>",
"sentence_end": "<ctx_sentence_end>",
"sentence_start": "<ctx_sentence_start>"
},
"default_boundary_after": "sentence_end",
"default_boundary_before": "sentence_start",
"emit_context_phone_tokens": false,
"linguistic_continuation_boundaries": [
"clause_continue"
],
"schema_version": 2
},
"emotion_guidance": {
"emotion_condition_mask_input": true,
"enabled": false,
"reference_condition_mask_input": true,
"strength_mixing": "runtime_linear_combination"
},
"emotions": {
"asset": "assets/emotions.json",
"count": 9,
"enabled": false
},
"energy": {
"default": 1.0,
"max": 1.5,
"min": 0.5
},
"fixed_shapes": {
"g2p_text_tokens": 512,
"latent_frames": 640,
"phone_frames": 512
},
"graph_input_contract": "scyllasband_affect_v1",
"pitch": {
"default": 0.0,
"max": 4.0,
"min": -4.0
},
"prefix_conditioning": {
"component": "vector_estimator",
"enabled": true,
"inputs": [
"prefix_latents",
"prefix_mask"
],
"max_frames": 24
},
"punctuation_silence": {
"silence_phone": "<sil>",
"target": "explicit_silence"
},
"reference_packs": {
"asset_dir": "assets/voice_packs",
"conditioning": {
"duration_predictor": "additive",
"vector_estimator": "additive",
"vocoder": "additive"
},
"enabled": true,
"fallback_weight": 0.25,
"manifest": "assets/voice_packs/reference_packs.json",
"normalization": "runtime_standardized_reference_pack_features",
"prosody_dim": 32,
"style_dim": 128
},
"runtime_acceleration": {
"backend": "litert",
"native_gpu_acceleration": {
"cuda_required": false,
"platform_targets": {
"android-arm64": {
"gpu_accelerator_library": "libLiteRtClGlAccelerator.so",
"gpu_backend": "OpenCL + OpenGL",
"runtime_library": "libLiteRt.so"
},
"android-x86_64": {
"gpu_accelerator_library": "libLiteRtClGlAccelerator.so",
"gpu_backend": "OpenCL + OpenGL",
"runtime_library": "libLiteRt.so"
},
"ios-arm64": {
"gpu_accelerator_library": "libLiteRtMetalAccelerator.dylib",
"gpu_backend": "Metal",
"runtime_library": "libLiteRt.dylib"
},
"ios-sim-arm64": {
"gpu_accelerator_library": "libLiteRtMetalAccelerator.dylib",
"gpu_backend": "Metal",
"runtime_library": "libLiteRt.dylib"
},
"linux-arm64": {
"gpu_accelerator_library": "libLiteRtWebGpuAccelerator.so",
"gpu_backend": "WebGPU (Vulkan)",
"runtime_library": "libLiteRt.so"
},
"linux-x86_64": {
"gpu_accelerator_library": "libLiteRtWebGpuAccelerator.so",
"gpu_backend": "WebGPU (Vulkan)",
"runtime_library": "libLiteRt.so"
},
"macos-arm64": {
"gpu_accelerator_library": "libLiteRtMetalAccelerator.dylib",
"gpu_backend": "Metal",
"runtime_library": "libLiteRt.dylib"
},
"windows-x86_64": {
"gpu_accelerator_library": "libLiteRtWebGpuAccelerator.dll",
"gpu_backend": "WebGPU (Direct3D)",
"runtime_library": "libLiteRt.dll"
}
}
},
"runtime_version": "2.1.5",
"source": "scyllasband_hf_staging_inferred_from_bundle",
"vector_estimator": {
"execution": "single_graph",
"min_gpu_split_vector_latent_frames": null,
"policy": "cpu_or_runtime_selected_single_graph",
"prefix_accelerator": null,
"split_artifacts": [],
"split_artifacts_available": false,
"split_block": null,
"tail_accelerator": null
},
"vocoder": {
"accelerator": "gpu_when_runtime_accelerator_gpu",
"fallback_accelerator": "cpu",
"policy": "gpu_vocoder_default_for_gpu_accelerator"
}
},
"sampler": {
"choices": [
"euler",
"heun"
],
"default": "heun"
},
"span_conditioning": {
"context_max_phones": 768,
"enabled": true,
"encoder_component": "vector_context_encoder",
"encoder_inputs": [
"span_context_phone_ids",
"span_context_segment_ids",
"span_context_mask"
],
"hidden_size": 512,
"segments": 3,
"solver_input": "span_context_hidden"
},
"speed": {
"default": 1.0,
"max": 1.25,
"min": 0.75
},
"steps": {
"default": 8,
"max": 64,
"min": 1
},
"target_buckets": {
"bucket_margin_frames": 0,
"buckets": [
{
"default": false,
"latent_frames": 256,
"vector_estimator": "vector_estimator_256",
"vector_estimator_prefix": "",
"vector_estimator_tail": "",
"vocoder": "vocoder_256"
},
{
"default": false,
"latent_frames": 384,
"vector_estimator": "vector_estimator_384",
"vector_estimator_prefix": "",
"vector_estimator_tail": "",
"vocoder": "vocoder_384"
},
{
"default": false,
"latent_frames": 512,
"vector_estimator": "vector_estimator_512",
"vector_estimator_prefix": "",
"vector_estimator_tail": "",
"vocoder": "vocoder_512"
},
{
"default": true,
"latent_frames": 640,
"vector_estimator": "vector_estimator",
"vector_estimator_prefix": "",
"vector_estimator_tail": "",
"vocoder": "vocoder"
}
],
"default_latent_frames": 640,
"enabled": true,
"policy": "smallest_fit"
}
},
"dataset_summary": {
"duration_hours": 1304.7728644444571,
"input_dir": "output/emotional_multilingual",
"languages": {
"en_gb": 89065,
"en_us": 209290,
"es": 294696,
"it": 293124
},
"max_duration_seconds": 36.4,
"min_duration_seconds": 1.04,
"samples": 886175,
"voices": 10
},
"default_language": "en_us",
"export_status": "complete_litert",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"model_name": "scyllasband",
"preferred_backends": [
"litert"
],
"voices": [
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/ariadne.npz",
"id": "ariadne",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "ariadne",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/felix.npz",
"id": "felix",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "felix",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/gwen.npz",
"id": "gwen",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "gwen",
"styles": [
"default"
]
},
{
"default_language": "en_gb",
"embedding_path": "assets/voice_packs/ink.npz",
"id": "ink",
"languages": [
"en_gb",
"en_us",
"es",
"it"
],
"name": "ink",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/max.npz",
"id": "max",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "max",
"styles": [
"default"
]
},
{
"default_language": "en_gb",
"embedding_path": "assets/voice_packs/orpheus.npz",
"id": "orpheus",
"languages": [
"en_gb",
"en_us",
"es",
"it"
],
"name": "orpheus",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/rex.npz",
"id": "rex",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "rex",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/scylla.npz",
"id": "scylla",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "scylla",
"styles": [
"default"
]
},
{
"default_language": "en_us",
"embedding_path": "assets/voice_packs/stone.npz",
"id": "stone",
"languages": [
"en_us",
"en_gb",
"es",
"it"
],
"name": "stone",
"styles": [
"default"
]
},
{
"default_language": "en_gb",
"embedding_path": "assets/voice_packs/tuesday.npz",
"id": "tuesday",
"languages": [
"en_gb",
"en_us",
"es",
"it"
],
"name": "tuesday",
"styles": [
"default"
]
}
]
}