Prev
# text baseline 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.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/projects-bucket-pack.json
    batch_size: 1
    gpus: 2
    inference:
      diff_steps: 400
      exp: 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: projects_pack_1/yt4_baseline_lats__2024-06-27_12-43-45
    ticket: TTS-388
  basket_generation_git_hash: 3ad347e2f731481af82dafbe72aa4daa136f907b
model_data_type: tts-cloning
ticket: TTS-388
version: 2024-06-27_12-43-45
encodec_mbr 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 2
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_encodec-opt-bigbatch
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_span: 50.0
        use_cache: true
      override_conditioning_features:
        c50: 0.0
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 6
        mode: MBR
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_dsat_mapping_encodec_mbrlat_override_optimizeloudness/yt4_langbycond_revgrad1_encodec-opt-bigbatch__2024-07-09_23-56-02
    ticket: TTS-393
  basket_generation_git_hash: c8fbcc9a0f18330789b7a4441a1efbeea1019935
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-07-09_23-56-02
encodec_mbr_24kfix 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_encodec-opt-bigbatch/
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 48000
      override_conditioning_features:
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 6
        mode: MBR
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_dsat_mapping_encodec_mbrlat_override_optimizeloudness/yt4_langbycond_revgrad1_encodec-opt-bigbatch__2024-07-22_14-21-18
    ticket: TTS-393
  basket_generation_git_hash: 37cb3542e38cc2685aebaf1d162ad5a0b7140f8e
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-07-22_14-21-18
encodec_mbr_24kfix-inhousediff-overrides 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      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_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_dsat_mapping_encodec_mbrlat_override_optimizeloudness_inhdiff/yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes__2024-08-02_03-20-48
    ticket: TTS-393
  basket_generation_git_hash: 5707d4c683a4b30e1858324363726a1e6d05bdd8
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-08-02_03-20-48
encodec_mbr_24kfix-normref 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_encodec-opt-bigbatch/
      force_reference_std: -23
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 48000
      override_conditioning_features:
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 6
        mode: MBR
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_dsat_mapping_encodec_mbrlat_override_optimizeloudness/yt4_langbycond_revgrad1_encodec-opt-bigbatch__2024-07-22_16-15-37
    ticket: TTS-393
  basket_generation_git_hash: 37cb3542e38cc2685aebaf1d162ad5a0b7140f8e
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-07-22_16-15-37
encodec_mbr_24kfix-inhousediff-overrides-normloud 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes
      force_reference_std: -23
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        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_dsat_mapping_encodec_mbrlat_override_optimizeloudness_inhdiff/yt4_langbycond_revgrad1_encodec-opt-bigbatch-diffcodes__2024-08-02_10-29-24
    ticket: TTS-393
  basket_generation_git_hash: 5707d4c683a4b30e1858324363726a1e6d05bdd8
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-08-02_10-29-24
encodec_mbr_normloud-contr-oversnr-dfnet 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud/
      force_reference_std: -21.5
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 48000
      override_conditioning_features:
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 6
        mode: MBR
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_dsat_mapping_encodec_mbrlat_override_optimizeloudness/yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud__2024-07-14_18-13-45
    ticket: TTS-393
  basket_generation_git_hash: c8fbcc9a0f18330789b7a4441a1efbeea1019935
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-07-14_18-13-45
encodec_mbr_normloud-contr-oversnr-dfnet-24kfix 
basket_config_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
data_meta: null
exp_name: yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud
lang: en
meta:
  basket_generation_config:
    basket_lang: en
    basket_path: quality/tts/tortoise-baskets/dsat_to_en_5projects_704_fixedref.json
    batch_size: 1
    gpus: 1
    inference:
      condition_sample_rate: 24000
      diff_on_codes: false
      diff_steps: 400
      exp: /mount/s3/tts-binary-data-nb/dimdi-y/yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud/
      force_reference_std: -21.5
      gpt_generate_args:
        do_sample: true
        num_return_sequences: 50
        prefix_allowed_tokens_fn: encodec_interleaved_layers
        repetition_penalty_span: 50.0
        use_cache: true
      out_sample_rate: 48000
      override_conditioning_features:
        snr: 100.0
      reranking_options:
        cdist_time_downsampling_factor: 6
        mode: MBR
        top_k: 1
      vocoder: none
    num_workers: 1
    output_dir: es_en_dsat_mapping_encodec_mbrlat_override_optimizeloudness/yt4_langbycond_revgrad1-contrastive_encodec-opt-bigbatch_normloud__2024-07-22_14-43-07
    ticket: TTS-393
  basket_generation_git_hash: 37cb3542e38cc2685aebaf1d162ad5a0b7140f8e
model_data_type: tts-cloning
ticket: TTS-393
version: 2024-07-22_14-43-07
120
From... Madrid
121
Madrid is very large.
122
What neighborhood am I from?
123
Well, I grew up in the Marqués de Salamanca neighborhood.
124
I told you.
125
My name is Adela.
126
Hello Adela, how are you?
127
Hello.
128
Pleased to meet you.
129
Well, you were working on the accented words, weren't you? The acute ones... There are the acute ones, there are the flat ones...
130
Hey!
131
Tell me.
132
And why doesn't Amparo continue?
133
Because... there's been a little change and... and this morning I'm going to stay.
134
But are you a teacher or what?
135
Yes, I am a teacher, I studied teaching.
136
May I have a moment, please?
137
The acute words, the... Please, just a moment... Just a moment... I understand... I understand that you prefer the other teacher because you don't know me at all...
138
I have just arrived.
139
But maybe I can show you something.
140
And I'm sure you can teach me something.
141
A lot.
142
I'm sure it is, I'm sure it is a lot.
143
But what I think is that... we have to do it together.
144
Us and those in the Salamanca neighborhood.
145
Well, be that as it may, I think the most important thing is that we respect each other.
146
Because if we do not respect ourselves, how can we pretend to be respectable to others?
147
I really feel that this way... we can all be much stronger.
148
And... alas...
149
Amelia! Come!
Next