Text-to-Speech
Pocket-TTS
Safetensors
Korean
audio
korean
voice-cloning
sentencepiece
knowledge-distillation
Instructions to use seastar105/pocket-tts-korean-100m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Pocket-TTS
How to use seastar105/pocket-tts-korean-100m with Pocket-TTS:
from pocket_tts import TTSModel import scipy.io.wavfile tts_model = TTSModel.load_model("seastar105/pocket-tts-korean-100m") voice_state = tts_model.get_state_for_audio_prompt( "hf://kyutai/tts-voices/alba-mackenna/casual.wav" ) audio = tts_model.generate_audio(voice_state, "Hello world, this is a test.") # Audio is a 1D torch tensor containing PCM data. scipy.io.wavfile.write("output.wav", tts_model.sample_rate, audio.numpy()) - Notebooks
- Google Colab
- Kaggle
Download eval/cv3_ema_scores.csv from seastar105/pocket-tts-korean-100m: direct link, hf CLI and curl.
- Browser
- Download file 5.01 kB
-
https://huggingface.co/seastar105/pocket-tts-korean-100m/resolve/main/eval/cv3_ema_scores.csv
- Command line
-
hf download hf://seastar105/pocket-tts-korean-100m/eval/cv3_ema_scores.csv
-
curl -L -o cv3_ema_scores.csv https://huggingface.co/seastar105/pocket-tts-korean-100m/resolve/main/eval/cv3_ema_scores.csv
5.01 kB
| model,step,checkpoint,cer,cer_no_space,speaker_similarity,utmos,silent,no_eos,num_items | |
| student,2000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00002000/model.safetensors,0.4898058038663085,0.5391802808827745,0.9186617337465286,2.7888201496601104,0,9,500 | |
| student,4000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00004000/model.safetensors,0.10062089920295918,0.09899684723416452,0.8905847424268722,2.42755991601944,0,0,500 | |
| student,6000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00006000/model.safetensors,0.07754634726319962,0.07750071653768988,0.887164558172226,2.369404767513275,0,0,500 | |
| student,8000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00008000/model.safetensors,0.08723413624554141,0.08615649183147034,0.887837360739708,2.3245253880023955,0,0,500 | |
| student,10000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00010000/model.safetensors,0.08032057774450659,0.08042419031241044,0.8880442825555801,2.3161566760540007,0,0,500 | |
| student,12000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00012000/model.safetensors,0.08133339204720595,0.08071080538836342,0.8867759425640106,2.307565588235855,0,0,500 | |
| student,14000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00014000/model.safetensors,0.08265445418116166,0.08403554026941817,0.888342763543129,2.2929549605846407,0,0,500 | |
| student,16000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00016000/model.safetensors,0.09322295125280726,0.093493837775867,0.8881259624958038,2.2944923810958864,0,0,500 | |
| student,18000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00018000/model.safetensors,0.08357919767493065,0.08466609343651477,0.8873635882139206,2.273546406745911,0,1,500 | |
| student,20000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00020000/model.safetensors,0.07983618829538949,0.07899111493264546,0.8857614450454712,2.2857840464115142,0,0,500 | |
| student,22000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00022000/model.safetensors,0.09115328724294333,0.09395242189739181,0.8870914011001587,2.2884030878543853,0,0,500 | |
| student,24000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00024000/model.safetensors,0.0983310581707693,0.1004299226139295,0.883941954612732,2.2648509922027587,0,0,500 | |
| student,26000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00026000/model.safetensors,0.0881588797393104,0.08879335053023789,0.885072045326233,2.251253634929657,0,0,500 | |
| student,28000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00028000/model.safetensors,0.09379541151085473,0.09309257666953281,0.8864236724376678,2.2606623768806458,0,0,500 | |
| student,30000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00030000/model.safetensors,0.09309084503941169,0.08896531957580969,0.8847276725769043,2.2566837058067324,0,0,500 | |
| student,32000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00032000/model.safetensors,0.08903958782861421,0.08770421324161651,0.8879240772724152,2.24206427526474,0,0,500 | |
| student,34000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00034000/model.safetensors,0.09648157118323132,0.09658928059615936,0.8844729942083359,2.250446535348892,0,0,500 | |
| student,36000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00036000/model.safetensors,0.09639350037430094,0.09509888220120379,0.8822739790678025,2.2466430897712706,0,1,500 | |
| student,38000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00038000/model.safetensors,0.08723413624554141,0.08741759816566351,0.8807653665542603,2.2279606750011443,0,4,500 | |
| student,40000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00040000/model.safetensors,0.09806684574397816,0.10243622814560045,0.8811250861287117,2.242311154603958,0,2,500 | |
| student,42000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00042000/model.safetensors,0.09771456250825664,0.09733447979363714,0.8794515370130539,2.2354571487903594,0,4,500 | |
| student,44000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00044000/model.safetensors,0.09088907481615219,0.09177414732014905,0.8823774423599243,2.2310054676532745,0,4,500 | |
| student,46000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00046000/model.safetensors,0.10991236954511427,0.11539122957867584,0.8808597807884216,2.21132421708107,0,2,500 | |
| student,48000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00048000/model.safetensors,0.09353119908406358,0.09292060762396102,0.8818871787786484,2.2253025748729707,0,2,500 | |
| student,50000,runs/korean_depth_distill_pretrained_init_4x5090_50k/checkpoints/step_00050000/model.safetensors,0.09974019111365538,0.10106047578102609,0.8825553715229034,2.2265217018127443,0,3,500 | |