Prev
# uttid text baseline
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_baseline_lats
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: /home/polovick/v2v_diff/ml/projects/ai-voice-cloning/dsat-basket-extended-refs-dur.json
    batch_size: 1
    gpus: 2
    inference:
      diff_steps: 400
      exp: /home/polovick/v2v_diff/ml/projects/ai-voice-cloning/yt4_baseline_lats
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        mode: MBR
        top_k: 1
      target_len_rate: 0.75
      vocoder: univnet
    num_workers: 1
    output_dir: dsat-cleared/yt4_baseline_lats__2024-07-30_03-28-45
    ticket: QUALITY-54
  basket_generation_git_hash: e0df79f1213deffbae77e909499694944e0746da
model_data_type: tts-cloning
ticket: QUALITY-54
version: 2024-07-30_03-28-45
encodec-inhousediff-sameinfer
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes__diff_nonorm_2codes_pretrained
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      diffusion_exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloudness_delaypattern8/diff_nonorm_2codes_pretrained
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_activation_span: 4.0
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 24000
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 2
        mode: MBR
        sakoe_chiba_radius: 24
        top_k: 1
      vocoder: bigvgan
    num_workers: 1
    output_dir: es_en_clean-dsat_mapping_encodec_mbrlat_inhdiff/yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes__diff_nonorm_2codes_pretrained__2024-09-18_17-36-55
    ticket: TTS-393
  basket_generation_git_hash: 75e464c6d886d92ef5b904f41695658ca2bc7545
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-09-18_17-36-55
xcodec160k_1code_inhdiff20k
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_xcodec-opt-bigbatch__diffusion_yt4_xcodec
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: true
      diff_steps: 400
      diffusion_exp: /mount/s3/tts-binary-data-nb/dimdi-y/diffusion_yt4_xcodec
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_xcodec-opt-bigbatch
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        repetition_penalty_activation_span: 4.0
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 24000
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        mode: MBR
        sakoe_chiba_radius: 16
        top_k: 1
      vocoder: bigvgan
    num_workers: 1
    output_dir: es_en_clean-dsat_mapping_xcodec_mbrlat_inhdiff/yt4_langbycond_revgrad1_xcodec-opt-bigbatch__diffusion_yt4_xcodec__2024-10-02_15-50-47
    ticket: TTS-393
  basket_generation_git_hash: 75e464c6d886d92ef5b904f41695658ca2bc7545
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-10-02_15-50-47
wavtokeniser
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: true
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        repetition_penalty_activation_span: 1.0
        repetition_penalty_span: 100.0
        use_cache: true
      out_sample_rate: 24000
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        mode: MBR
        sakoe_chiba_radius: 24
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_clean-dsat_mapping_xcodec_mbrlat_codecdec/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc__2024-10-29_09-42-52
    ticket: TTS-392
  basket_generation_git_hash: 75e464c6d886d92ef5b904f41695658ca2bc7545
model_data_type: tts-cloning
ticket: TTS-392
version: 2024-10-29_09-42-52
wavtokeniser_normloud
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: true
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
      force_reference_std: -23
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        repetition_penalty_activation_span: 1.0
        repetition_penalty_span: 100.0
        use_cache: true
      out_sample_rate: 24000
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        mode: MBR
        sakoe_chiba_radius: 24
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_clean-dsat_mapping_xcodec_mbrlat_codecdec/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc__2024-10-29_09-58-08
    ticket: TTS-392
  basket_generation_git_hash: 75e464c6d886d92ef5b904f41695658ca2bc7545
model_data_type: tts-cloning
ticket: TTS-392
version: 2024-10-29_09-58-08
wavtokeniser_normloud_lesspenalty
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_cleared_721.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: true
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc
      force_reference_std: -23
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        repetition_penalty: 1.5
        repetition_penalty_activation_span: 5.0
        repetition_penalty_span: 100.0
        use_cache: true
      out_sample_rate: 24000
      override_conditioning_features:
        c50: 0.0
        pitch_std: 100.0
        snr: 100.0
      reranking_options:
        mode: MBR
        sakoe_chiba_radius: 24
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_clean-dsat_mapping_xcodec_mbrlat_codecdec/yt4_langbycond_revgrad1_wavtokenizer-opt-bigbatch_t5enc__2024-10-29_13-45-07
    ticket: TTS-392
  basket_generation_git_hash: 75e464c6d886d92ef5b904f41695658ca2bc7545
model_data_type: tts-cloning
ticket: TTS-392
version: 2024-10-29_13-45-07
420
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0036
We promised each other to try hard, to back each other up, to keep each other out of trouble.
421
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0037
Kat, I am here.
422
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0038
This is precisely the question.
423
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0039
That night I found something.
424
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0040
Strange, isn't it?
425
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0041
The chick who dropped her...pays only cash but doesn't work and is always alone.
426
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0042
I also looked for something on the Internet, but found only an old profile and some bullshit about the coin.
427
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0043
Wait, wait.
428
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0044
Are you telling me you've been stalking a chick, Cat?
429
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0045
So you don't make me feel good at all.
430
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0046
What is this?
431
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0047
Look, in my opinion this chick is into something shady.
432
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0048
I can feel it.
433
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0049
If it was something big, I need to know!
434
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0050
Imagine a video about it.
435
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0051
At least get help, right?
436
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0052
We are in this together.
437
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0053
No, Mika.
438
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0054
Tonight I have a date with her.
439
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0055
Beautifully romantic, under the moon, I have to go alone.
440
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0056
Come on, don't joke.
441
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0057
If this is really inside something big, as you say, it could be dangerous.
442
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0058
I'm not kidding, Mika.
443
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0059
But don't worry, I'll stream everything and that way if something happens, you can come and cover my back, right?
444
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0060
Don't worry.
445
LA6_Full_ProresHQ-2-_it_F0/LA6_Full_ProresHQ-2-_it_F0_0061
Look, Kat, listen to me...
446
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0062
No, please, Mika.
447
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0063
Just leave me alone, okay?
448
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0064
Involving you now would do no good.
449
LA6_Full_ProresHQ-2-_it_F1/LA6_Full_ProresHQ-2-_it_F1_0065
I'll call you, okay?
Next