asr

kaztts_male2_tacotron2_train.loss.ave/exp/tts_stats_raw_char/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dca35226cf181d840baedf98032bea16d3cb4b69496cd59adb8f04b49d298fc
+size 1402

kaztts_male2_tacotron2_train.loss.ave/exp/tts_train_raw_char/config.yaml

@@ -0,0 +1,231 @@
+config: conf/train.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_train_raw_char
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: 0
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 200
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 20
+valid_batch_size: null
+batch_bins: 5120000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_char/train/text_shape.char
+- exp/tts_stats_raw_char/train/speech_shape
+valid_shape_file:
+- exp/tts_stats_raw_char/valid/text_shape.char
+- exp/tts_stats_raw_char/valid/speech_shape
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 150
+- 204800
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/tr_no_dev/wav.scp
+    - speech
+    - sound
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 0.001
+    eps: 1.0e-06
+    weight_decay: 0.0
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- <space>
+- а
+- ы
+- е
+- н
+- т
+- р
+- і
+- л
+- с
+- д
+- қ
+- м
+- к
+- о
+- б
+- ж
+- у
+- з
+- и
+- ғ
+- п
+- ң
+- ш
+- й
+- г
+- .
+- ү
+- ұ
+- ө
+- ','
+- ә
+- '-'
+- я
+- в
+- х
+- ц
+- ф
+- э
+- ь
+- ю
+- ч
+- ':'
+- '?'
+- ;
+- ъ
+- һ
+- '!'
+- щ
+- ё
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 1024
+    hop_length: 256
+    win_length: null
+    fs: 22050
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/tts_stats_raw_char/train/feats_stats.npz
+tts: tacotron2
+tts_conf:
+    embed_dim: 512
+    elayers: 1
+    eunits: 512
+    econv_layers: 3
+    econv_chans: 512
+    econv_filts: 5
+    atype: location
+    adim: 512
+    aconv_chans: 32
+    aconv_filts: 15
+    cumulate_att_w: true
+    dlayers: 2
+    dunits: 1024
+    prenet_layers: 2
+    prenet_units: 256
+    postnet_layers: 5
+    postnet_chans: 512
+    postnet_filts: 5
+    output_activation: null
+    use_batch_norm: true
+    use_concate: true
+    use_residual: false
+    dropout_rate: 0.5
+    zoneout_rate: 0.1
+    reduction_factor: 1
+    spk_embed_dim: null
+    use_masking: true
+    bce_pos_weight: 5.0
+    use_guided_attn_loss: true
+    guided_attn_loss_sigma: 0.4
+    guided_attn_loss_lambda: 1.0
+pitch_extract: null
+pitch_extract_conf: {}
+pitch_normalize: null
+pitch_normalize_conf: {}
+energy_extract: null
+energy_extract_conf: {}
+energy_normalize: null
+energy_normalize_conf: {}
+required:
+- output_dir
+- token_list
+version: 0.10.3a4
+distributed: false

kaztts_male2_tacotron2_train.loss.ave/exp/tts_train_raw_char/train.loss.ave_5best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:deb9cb49703b012d8f1406d15f4fa182b39ecb320bc4e2be5a557ac58766ea75
+size 106809178

kaztts_male2_tacotron2_train.loss.ave/meta.yaml

@@ -0,0 +1,9 @@
+espnet: 0.10.3a4
+files:
+  model_file: exp/tts_train_raw_char/train.loss.ave_5best.pth
+python: "3.8.12 | packaged by conda-forge | (default, Oct 12 2021, 21:59:51) \n[GCC\
+  \ 9.4.0]"
+timestamp: 1641919865.515724
+torch: 1.7.0
+yaml_files:
+  train_config: exp/tts_train_raw_char/config.yaml

parallelwavegan_male2_checkpoint/checkpoint-400000steps.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb415182fdf84dbe6a0bbca72e5ed9ed39009b622cf39e20ba630cb830594329
+size 17694194

parallelwavegan_male2_checkpoint/config.yml

@@ -0,0 +1,104 @@
+allow_cache: true
+batch_max_steps: 25600
+batch_size: 6
+config: conf/parallel_wavegan.v1.yaml
+dev_dumpdir: dump/dev/norm
+dev_feats_scp: null
+dev_segments: null
+dev_wav_scp: null
+discriminator_grad_norm: 1
+discriminator_optimizer_params:
+  eps: 1.0e-06
+  lr: 5.0e-05
+  weight_decay: 0.0
+discriminator_params:
+  bias: true
+  conv_channels: 64
+  in_channels: 1
+  kernel_size: 3
+  layers: 10
+  nonlinear_activation: LeakyReLU
+  nonlinear_activation_params:
+    negative_slope: 0.2
+  out_channels: 1
+  use_weight_norm: true
+discriminator_scheduler_params:
+  gamma: 0.5
+  step_size: 200000
+discriminator_train_start_steps: 100000
+distributed: false
+eval_interval_steps: 1000
+fft_size: 1024
+fmax: 7600
+fmin: 80
+format: hdf5
+generator_grad_norm: 10
+generator_optimizer_params:
+  eps: 1.0e-06
+  lr: 0.0001
+  weight_decay: 0.0
+generator_params:
+  aux_channels: 80
+  aux_context_window: 2
+  dropout: 0.0
+  gate_channels: 128
+  in_channels: 1
+  kernel_size: 3
+  layers: 30
+  out_channels: 1
+  residual_channels: 64
+  skip_channels: 64
+  stacks: 3
+  upsample_net: ConvInUpsampleNetwork
+  upsample_params:
+    upsample_scales:
+    - 4
+    - 4
+    - 4
+    - 4
+  use_weight_norm: true
+generator_scheduler_params:
+  gamma: 0.5
+  step_size: 200000
+global_gain_scale: 1.0
+hop_size: 256
+lambda_adv: 4.0
+log_interval_steps: 100
+num_mels: 80
+num_save_intermediate_results: 4
+num_workers: 2
+outdir: exp/train_nodev_parallel_wavegan.v1
+pin_memory: true
+pretrain: ''
+rank: 0
+remove_short_samples: true
+resume: ''
+sampling_rate: 22050
+save_interval_steps: 5000
+stft_loss_params:
+  fft_sizes:
+  - 1024
+  - 2048
+  - 512
+  hop_sizes:
+  - 120
+  - 240
+  - 50
+  win_lengths:
+  - 600
+  - 1200
+  - 240
+  window: hann_window
+train_dumpdir: dump/train_nodev/norm
+train_feats_scp: null
+train_max_steps: 400000
+train_segments: null
+train_wav_scp: null
+trim_frame_size: 2048
+trim_hop_size: 512
+trim_silence: false
+trim_threshold_in_db: 60
+verbose: 1
+version: 0.4.8
+win_length: null
+window: hann

requirements.txt

@@ -4,4 +4,7 @@ torchaudio
 transformers
 numpy
 scipy==1.13.1
-umsc
+umsc
+parallel_wavegan==0.6.1
+espnet==202412
+espnet-tts-frontend==0.0.3

tts.py

@@ -1,8 +1,13 @@
 from transformers import VitsModel, AutoTokenizer
 import torch
 import scipy.io.wavfile
+from parallel_wavegan.utils import load_model
+from espnet2.bin.tts_inference import Text2Speech
+from turkicTTS_utils import normalization
 import util
 
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
 # Load processor and model
 models_info = {
     "Meta-MMS": {
@@ -10,10 +15,37 @@ models_info = {
         "model": VitsModel.from_pretrained("facebook/mms-tts-uig-script_arabic"),
         "arabic_script": True
     },
+    "IS2AI-TurkicTTS": None
 }
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+vocoder_checkpoint="parallelwavegan_male2_checkpoint/checkpoint-400000steps.pkl" ### specify vocoder path
+vocoder = load_model(vocoder_checkpoint).to(device).eval()
+vocoder.remove_weight_norm()
+
+### specify path to the main model(transformer/tacotron2/fastspeech) and its config file
+config_file = "exp/tts_train_raw_char/config.yaml"
+model_path = "exp/tts_train_raw_char/train.loss.ave_5best.pth"
+
+text2speech = Text2Speech(
+    config_file,
+    model_path,
+    device=device, ## if cuda not available use cpu
+    ### only for Tacotron 2
+    threshold=0.5,
+    minlenratio=0.0,
+    maxlenratio=10.0,
+    use_att_constraint=True,
+    backward_window=1,
+    forward_window=3,
+    ### only for FastSpeech & FastSpeech2
+    speed_control_alpha=1.0,
+)
+text2speech.spc2wav = None  ### disable griffin-lim
 
 def synthesize(text, model_id):
+    if model_id == 'IS2AI-TurkicTTS':
+        return synthesize_turkic_tts(text)
+    
     if models_info[model_id]["arabic_script"]:
         text = util.ug_latn_to_arab(text)
     processor = models_info[model_id]["processor"]
@@ -27,4 +59,17 @@ def synthesize(text, model_id):
     sample_rate = model.config.sampling_rate
     scipy.io.wavfile.write(output_path, rate=sample_rate, data=output.numpy()[0])
 
-    return output_path
+    return output_path
+
+def synthesize_turkic_tts(text):
+    text = normalization(text, 'uyghur')
+    
+    with torch.no_grad():
+        c_mel = text2speech(text)['feat_gen']
+        wav = vocoder.inference(c_mel)
+    
+    output = wav.view(-1).cpu()
+
+    output_path = "tts_output.wav"
+    scipy.io.wavfile.write(output_path, rate=22050, data=output.numpy()[0])
+

turkicTTS_ipa_convert.py

@@ -0,0 +1,1815 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+
+'''
+2022.06.03
+kazakh_to_ipa() <> ipa_to_kazakh()
+test_kazakh()
+turkish_to_ipa() <> ipa_to_turkish()
+test_turkish()
+
+2022.07.05
+kyrgyz_to_ipa() <> ipa_to_kyrgyz()
+test_kyrgyz()
+uzbek_to_ipa() <> ipa_to_uzbek()
+test_uzbek()
+azerbaijani_to_ipa() <> ipa_to_azerbaijani()
+test_azerbaijani()
+turkmen_to_ipa() <> ipa_to_turkmen()
+test_turkmen()
+
+2022.07.07
+tatar_to_ipa() <> ipa_to_tatar()
+test_tatar()
+bashkir_to_ipa() <> ipa_to_bashkir()
+test_bashkir()
+sakha_to_ipa() <> ipa_to_sakha()
+test_sakha()
+
+2022.07.12
+experimentally added î and â to turkish_to_ipa()
+
+2022.08.04
+uyghur_to_ipa() <> ipa_to_uyghur()
+'''
+
+import re
+    
+
+# kazakh scripts
+
+def kazakh_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list.
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # three-sound convenience vowels:
+    text = re.sub("[Юю]", "ǔ", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("[Цц]", "š", text)
+    text = re.sub("[Чч]", "ʆ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Яя]", "ǎ", text)
+    text = re.sub("[Ее]", "ě", text)
+    text = re.sub("[Ёё]", "ǒ", text)
+    text = re.sub("[Ии]", "ǐ", text)
+    text = re.sub("[Уу]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Бб]", "b", text)
+    text = re.sub("[Вв]", "v", text)
+    text = re.sub("[Гг]", "g", text)
+    text = re.sub("[Ғғ]", "ɣ", text)
+    text = re.sub("[Дд]", "d", text)
+    text = re.sub("[Жж]", "ʒ", text)
+    text = re.sub("[Зз]", "z", text)
+    text = re.sub("[Йй]", "j", text)
+    text = re.sub("[Кк]", "k", text)
+    text = re.sub("[Ққ]", "q", text)
+    text = re.sub("[Лл]", "l", text)
+    text = re.sub("[Мм]", "m", text)
+    text = re.sub("[Нн]", "n", text)
+    text = re.sub("[Ңң]", "ŋ", text)
+    text = re.sub("[Пп]", "p", text)
+    text = re.sub("[Рр]", "r", text)
+    text = re.sub("[Сс]", "s", text)
+    text = re.sub("[Тт]", "t", text)
+    text = re.sub("[Фф]", "f", text)
+    text = re.sub("[Хх]", "x", text)
+    text = re.sub("[Һһ]", "h", text)
+    text = re.sub("[Шш]", "ʃ", text)
+    text = re.sub("[Щщ]", "ɕ", text)
+    text = re.sub("[Ъъ]", "ʔ", text)
+    text = re.sub("[Ьь]", "ʲ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Аа]", "ɑ", text)
+    text = re.sub("[Әә]", "æ", text)
+    text = re.sub("[Оо]", "ɔ", text)
+    text = re.sub("[Өө]", "ɵ", text)
+    text = re.sub("[Ұұ]", "ʊ", text)
+    text = re.sub("[Үү]", "ʏ", text)
+    text = re.sub("[Ыы]", "ɤ", text)
+    text = re.sub("[Іі]", "ɪ", text)
+    text = re.sub("[Ээ]", "e", text)
+
+    # rules
+
+    '''
+    rule 1:
+    if [æ], [ě], [ɵ], [ʏ], [ɪ] are followed by [l] and [l] is NOT followed by [æ], [ě], [ɵ], [ʏ], [ɪ], or [ʲ],
+    use [ł] instead of [l] (e.g., [kěł], [kěłdɪ], but [kělěmɪn], [marsělʲ]).
+    '''
+    text = re.sub(r"([æěɵʏɪ])(l)([^æěɵʏɪʲ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letters "о" and "ө", [ɔ] and [ɵ] at the beginning of a word are preceded by [w] (e.g., осы [wɔsɤ] not [ɔsɤ], өзі [wɵzɪ] not [ɵzɪ]).
+    '''
+    text = re.sub(r"\b([ɔɵ])", r"w\1", text)
+
+    '''
+    rule 3
+    if the letter "у" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆʲ])", r"w\1", text)
+
+    '''
+    rule 4:
+    if the letter "и" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆʲ])", r"i\1", text)
+
+    return text
+
+def ipa_to_kazakh(text):
+    # three-sound convenience vowels:
+    text = re.sub("ǔ", "ю", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("š", "ц", text)
+    text = re.sub("ʆ", "ч", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ǎ", "я", text)
+    text = re.sub("ě", "е", text)
+    text = re.sub("ǒ", "ё", text)
+    text = re.sub("ǐ", "и", text)
+    text = re.sub("u", "у", text)
+
+    # single-sound consonants:
+    text = re.sub("b", "б", text)
+    text = re.sub("v", "в", text)
+    text = re.sub("g", "г", text)
+    text = re.sub("ɣ", "ғ",  text)
+    text = re.sub("d", "д", text)
+    text = re.sub("ʒ", "ж", text)
+    text = re.sub("z", "з", text)
+    text = re.sub("j", "й", text)
+    text = re.sub("k", "к", text)
+    text = re.sub("q", "қ", text)
+    text = re.sub("l", "л", text)
+    text = re.sub("m", "м", text)
+    text = re.sub("n", "н", text)
+    text = re.sub("ŋ", "ң", text)
+    text = re.sub("p", "п", text)
+    text = re.sub("r", "р", text)
+    text = re.sub("s", "с", text)
+    text = re.sub("t", "т", text)
+    text = re.sub("f", "ф", text)
+    text = re.sub("x", "х", text)
+    text = re.sub("h", "һ", text)
+    text = re.sub("ʃ", "ш", text)
+    text = re.sub("ɕ", "щ", text)
+    text = re.sub("ʔ", "ъ", text)
+    text = re.sub("ʲ", "ь", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "а", text)
+    text = re.sub("æ", "ә", text)
+    text = re.sub("ɔ", "о", text)
+    text = re.sub("ɵ", "ө", text)
+    text = re.sub("ʊ", "ұ", text)
+    text = re.sub("ʏ", "ү", text)
+    text = re.sub("ɤ", "ы", text)
+    text = re.sub("ɪ", "і", text)
+    text = re.sub("e", "э", text)
+
+    # anti-rules
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([әеөүі])(ł)([^әеөүіь])", r"\1л\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"\bw([оө])", r"\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"w([бвгғджзйкқлмнңпрстфхһцчшщъьчц])", r"у\1", text)
+
+    '''
+    anti-rule 4:
+    the symbol [i] is used in one case only, so we can just replace it for и.
+    '''
+    text = re.sub(r"i", r"и", text)
+
+    '''
+    anti-rules for Turkish and Kyrgyz Ǯ, Turkish ł, azerbaijani ḡ, sakha ɲ
+    '''
+    text = re.sub(r"w([Ǯ])", r"у\1", text)
+    text = re.sub(r"Ǯ", r"дж", text)
+    text = re.sub(r"ł", r"ль", text)
+    text = re.sub(r"ḡ", r"гь", text)
+    text = re.sub(r"ɲ", r"нь", text)
+
+    return text
+
+# testing kazakh scripts
+
+def test_kazakh(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_kazakh(kazakh_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# turkish scripts
+
+def turkish_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list.
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # two-sound convenience consonants:
+    text = re.sub("[Cc]", "Ǯ", text)
+    text = re.sub("[Çç]", "ʆ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[İi]", "ǐ", text)
+    text = re.sub("[Uu]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Jj]", "ʒ", text)
+    text = re.sub("[Yy]", "j", text)
+    text = re.sub("[Bb]", "b", text)
+    text = re.sub("[Dd]", "d", text)
+    text = re.sub("[Ff]", "f", text)
+    text = re.sub("[Gg]", "g", text)
+    text = re.sub("[Ğğ]", "ɣ", text)
+    text = re.sub("[Hh]", "h", text)
+    text = re.sub("[Kk]", "k", text)
+    text = re.sub("[Ll]", "l", text)
+    text = re.sub("[Mm]", "m", text)
+    text = re.sub("[Nn]", "n", text)
+    text = re.sub("[Pp]", "p", text)
+    text = re.sub("[Rr]", "r", text)
+    text = re.sub("[Ss]", "s", text)
+    text = re.sub("[Şş]", "ʃ", text)
+    text = re.sub("[Tt]", "t", text)
+    text = re.sub("[Vv]", "v", text)
+    text = re.sub("[Zz]", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("[Aa]", "ɑ", text)
+    text = re.sub("[Ee]", "e", text)
+    text = re.sub("[Iı]", "ɤ", text)
+    text = re.sub("[Oo]", "ɔ", text)
+    text = re.sub("[Öö]", "ɵ", text)
+    text = re.sub("[Üü]", "ʏ", text)
+    text = re.sub("[Îî]", "ǐ", text) # experimentally added
+    text = re.sub("[Ââ]", "ɑ", text) # experimentally added
+
+
+    '''
+    rule 1:
+    if [e], [ɵ], [ʏ], [i] are followed by [l] and [l] is NOT followed by [e], [ɵ], [ʏ], or [i],
+    use [ł] instead of [l] (e.g., [geł], [gełdi], but [gelecek]).
+    '''
+    text = re.sub(r"([eɵʏǐ])(l)([^eɵʏǐ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letter "u" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdʒzklłmnprstfhʃʆǮ])", r"w\1", text)
+
+    '''
+    rule 3:
+    if the letter "i" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdʒzklłmnprstfhʃʆǮ])", r"i\1", text)
+
+    return text
+
+def ipa_to_turkish(text):
+    # two-sound convenience consonants:
+    text = re.sub("Ǯ", "c", text)
+    text = re.sub("ʆ", "ç", text)
+
+    # single-sound consonants:
+    text = re.sub("j", "y", text)
+    text = re.sub("ʒ", "j", text)
+    text = re.sub("b", "b", text)
+    text = re.sub("d", "d", text)
+    text = re.sub("f", "f", text)
+    text = re.sub("g", "g", text)
+    text = re.sub("ɣ", "ğ", text)
+    text = re.sub("h", "h", text)
+    text = re.sub("k", "k", text)
+    text = re.sub("l", "l", text)
+    text = re.sub("m", "m", text)
+    text = re.sub("n", "n", text)
+    text = re.sub("p", "p", text)
+    text = re.sub("r", "r", text)
+    text = re.sub("s", "s", text)
+    text = re.sub("ʃ", "ş", text)
+    text = re.sub("t", "t", text)
+    text = re.sub("v", "v", text)
+    text = re.sub("z", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "a", text)
+    text = re.sub("e", "e", text)
+    text = re.sub("ɤ", "ı", text)
+    text = re.sub("ǐ", "i", text)
+    text = re.sub("ɔ", "o", text)
+    text = re.sub("ɵ", "ö", text)
+    text = re.sub("ʏ", "ü", text)
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([eöüi])(ł)([^eöüi])", r"\1l\3", text)
+
+    '''
+    anti-rule 2:
+    the symbol [w] is used in one case only, so we can just replace it for u.
+    '''
+    text = re.sub(r"w", r"u", text)
+
+    '''
+    anti-rule 3:
+    the symbol [i] is used in one case only, so we can just replace it for i.
+    '''
+    text = re.sub(r"i", r"i", text)
+
+    return text
+
+# testing turkish scripts
+
+def test_turkish(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_turkish(turkish_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# kyrgyz scripts
+
+def kyrgyz_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # three-sound convenience vowels:
+    text = re.sub("[Юю]", "ǔ", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("[Цц]", "š", text)
+    text = re.sub("[Чч]", "ʆ", text)
+    text = re.sub("[Жж]", "Ǯ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Яя]", "ǎ", text)
+    text = re.sub("[Ее]", "ě", text)
+    text = re.sub("[Ёё]", "ǒ", text)
+    text = re.sub("[Ии]", "ǐ", text)
+    text = re.sub("[Уу]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Бб]", "b", text)
+    text = re.sub("[Вв]", "v", text)
+    text = re.sub("[Гг]", "g", text)
+    text = re.sub("[Дд]", "d", text)
+    text = re.sub("[Зз]", "z", text)
+    text = re.sub("[Йй]", "j", text)
+    text = re.sub("[Кк]", "k", text)
+    text = re.sub("[Лл]", "l", text)
+    text = re.sub("[Мм]", "m", text)
+    text = re.sub("[Нн]", "n", text)
+    text = re.sub("[Ңң]", "ŋ", text)
+    text = re.sub("[Пп]", "p", text)
+    text = re.sub("[Рр]", "r", text)
+    text = re.sub("[Сс]", "s", text)
+    text = re.sub("[Тт]", "t", text)
+    text = re.sub("[Фф]", "f", text)
+    text = re.sub("[Хх]", "x", text)
+    text = re.sub("[Шш]", "ʃ", text)
+    text = re.sub("[Щщ]", "ɕ", text)
+    text = re.sub("[Ъъ]", "ʔ", text)
+    text = re.sub("[Ьь]", "ʲ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Аа]", "ɑ", text)
+    text = re.sub("[Оо]", "ɔ", text)
+    text = re.sub("[Өө]", "ɵ", text)
+    text = re.sub("[Үү]", "ʏ", text)
+    text = re.sub("[Ыы]", "ɤ", text)
+    text = re.sub("[Ээ]", "e", text)
+
+    # rules 1-4 are similar to those for Kazakh:
+
+    '''
+    rule 1:
+    if [ě], [ɵ], [ʏ], are followed by [l] and [l] is NOT followed by [ě], [ɵ], [ʏ], or [ʲ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([ɵʏě])(l)([^ɵʏěʲ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letters "о" and "ө", [ɔ] and [ɵ] at the beginning of a word are followed by [w] (e.g., осы [wɔsɤ] not [ɔsɤ], өзі [wɵzɪ] not [ɵzɪ]).
+    '''
+    text = re.sub(r"\b([ɔɵ])", r"w\1", text)
+
+    '''
+    rule 3
+    if the letter "у" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdzjkqlłmnŋprstfxhʃɕʔšʆǮʲ])", r"w\1", text)
+
+    '''
+    rule 4:
+    if the letter "и" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdzjkqlłmnŋprstfxhʃɕʔšʆǮʲ])", r"i\1", text)
+
+    # rules 5-6 are specific to Kyrgyz:
+
+    '''
+    rule 5
+    ɑ|ɔ|u|ɤ + k + ɑ|ɔ|u|ɤ
+    '''
+    text = re.sub(r"([ɑɔwɤ])k", r"\1q", text)
+    text = re.sub(r"k([ɑɔuɤ])", r"q\1", text)
+
+    '''
+    rule 6
+    ɑ|��|u|ɤ + g + ɑ|ɔ|u|ɤ
+    '''
+    text = re.sub(r"([ɑɔwɤ])g", r"\1ɣ", text)
+    text = re.sub(r"g([ɑɔuɤ])", r"ɣ\1", text)
+
+    return text
+
+def ipa_to_kyrgyz(text):
+    # three-sound convenience vowels:
+    text = re.sub("ǔ", "ю", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("š", "ц", text)
+    text = re.sub("ʆ", "ч", text)
+    text = re.sub("Ǯ", "ж", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ǎ", "я", text)
+    text = re.sub("ě", "е", text)
+    text = re.sub("ǒ", "ё", text)
+    text = re.sub("ǐ", "и", text)
+    text = re.sub("u", "у", text)
+
+    # single-sound consonants:
+    text = re.sub("b", "б", text)
+    text = re.sub("v", "в", text)
+    text = re.sub("g", "г", text)
+    text = re.sub("ɣ", "г", text)
+    text = re.sub("d", "д", text)
+    text = re.sub("z", "з", text)
+    text = re.sub("j", "й", text)
+    text = re.sub("k", "к", text)
+    text = re.sub("l", "л", text)
+    text = re.sub("m", "м", text)
+    text = re.sub("n", "н", text)
+    text = re.sub("ŋ", "ң", text)
+    text = re.sub("p", "п", text)
+    text = re.sub("q", "к", text)
+    text = re.sub("r", "р", text)
+    text = re.sub("s", "с", text)
+    text = re.sub("t", "т", text)
+    text = re.sub("f", "ф", text)
+    text = re.sub("x", "х", text)
+    text = re.sub("ʃ", "ш", text)
+    text = re.sub("ɕ", "щ", text)
+    text = re.sub("ʔ", "ъ", text)
+    text = re.sub("ʲ", "ь", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "а", text)
+    text = re.sub("ɔ", "о", text)
+    text = re.sub("ɵ", "ө", text)
+    text = re.sub("ʏ", "ү", text)
+    text = re.sub("ɤ", "ы", text)
+    text = re.sub("e", "э", text)
+
+    # anti-rules 1-4 are similar to those for Kazakh:
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([өүе])(ł)([^өүеʲ])", r"\1л\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"\bw([оө])", r"\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"w([бвгдзйклмнңпрстфхцчшщъьчцж])", r"у\1", text)
+
+    '''
+    anti-rule 4:
+    '''
+    text = re.sub(r"i([бвгдзйклмнңпрстфхцчшщъьчцж])", r"и\1", text)
+
+    return text
+
+# testing kyrgyz scripts
+
+def test_kyrgyz(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_kyrgyz(kyrgyz_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# uzbek scripts
+
+def uzbek_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # two-sound convenience consonants:
+    text = re.sub("[Jj]", "Ǯ", text)
+    text = re.sub("Ch", "ʆ", text)
+    text = re.sub("ch", "ʆ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Ii]", "ǐ", text)
+    text = re.sub("[Uu]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Bb]", "b", text)
+    text = re.sub("[Dd]", "d", text)
+    text = re.sub("[Ff]", "f", text)
+    text = re.sub("G‘", "ɣ", text)
+    text = re.sub("g‘", "ɣ", text)
+    text = re.sub("[Gg]", "g", text)
+    text = re.sub("[Hh]", "h", text)
+    text = re.sub("[Kk]", "k", text)
+    text = re.sub("[Ll]", "l", text)
+    text = re.sub("[Mm]", "m", text)
+    text = re.sub("[Nn]", "n", text)
+    text = re.sub("Ng", "ŋ", text)
+    text = re.sub("ng", "ŋ", text)
+    text = re.sub("[Pp]", "p", text)
+    text = re.sub("[Qq]", "q", text)
+    text = re.sub("[Rr]", "r", text)
+    text = re.sub("[Ss]", "s", text)
+    text = re.sub("Sh", "ʃ", text)
+    text = re.sub("sh", "ʃ", text)
+    text = re.sub("[Tt]", "t", text)
+    text = re.sub("[Vv]", "v", text)
+    text = re.sub("[Xx]", "x", text)
+    text = re.sub("[Yy]", "j", text)
+    text = re.sub("[Zz]", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("[Aa]", "æ", text)
+    text = re.sub("[Ee]", "e", text)
+    text = re.sub("Oʻ", "ɵ", text)
+    text = re.sub("oʻ", "ɵ", text)
+    text = re.sub("[Oo]", "ɔ", text)
+
+    # hard sign
+    text = re.sub("'", "ʔ", text)
+
+    '''
+    rule 1:
+    if [æ], [e], [ɵ], [ǐ] are followed by [l] and [l] is NOT followed by [æ], [e], [ɵ], [ǐ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æɵǐe])(l)([^æɵǐe])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letter "u" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdzjkqlłmnŋprstfxhʃʔʆǮ])", r"w\1", text)
+
+    '''
+    rule 3:
+    if the letter "i" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdzjkqlłmnŋprstfxhʃʔʆǮ])", r"i\1", text)
+
+    return text
+
+def ipa_to_uzbek(text):
+    # two-sound convenience consonants:
+    text = re.sub("j", "y", text)  # exception! precedence issue
+    text = re.sub("Ǯ", "j", text)
+    text = re.sub("ʆ", "ch", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ǐ", "i", text)
+    text = re.sub("u", "u", text)
+
+    # single-sound convenience consonants:
+    text = re.sub("b", "b", text)
+    text = re.sub("d", "d", text)
+    text = re.sub("f", "f", text)
+    text = re.sub("g", "g", text)
+    text = re.sub("ɣ", "g‘", text)
+    text = re.sub("h", "h", text)
+    text = re.sub("k", "k", text)
+    text = re.sub("l", "l", text)
+    text = re.sub("m", "m", text)
+    text = re.sub("n", "n", text)
+    text = re.sub("ŋ", "ng", text)
+    text = re.sub("p", "p", text)
+    text = re.sub("q", "q", text)
+    text = re.sub("r", "r", text)
+    text = re.sub("s", "s", text)
+    text = re.sub("ʃ", "sh", text)
+    text = re.sub("t", "t", text)
+    text = re.sub("v", "v", text)
+    text = re.sub("x", "x", text)
+    text = re.sub("z", "z", text)
+
+    # single-sound convenience vowels:
+    text = re.sub("æ", "a", text)
+    text = re.sub("e", "e", text)
+    text = re.sub("ɵ", "o‘", text)
+    text = re.sub("ɔ", "o", text)
+
+    # hard sign
+    text = re.sub("ʔ", "'", text)
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([aei‘])(ł)([^aei‘])", r"\1l\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"w([bcvgɣdjzklmnpqrstfhyx])", r"u\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"i([bcvgɣdjzklmnpqrstfhyx])", r"i\1", text)
+
+    return text
+
+# testing uzbek scripts
+
+def test_uzbek(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_uzbek(uzbek_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# azerbaijani scripts
+
+def azerbaijani_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # two-sound convenience consonants:
+    text = re.sub("[Cc]", "Ǯ", text)
+    text = re.sub("[Çç]", "ʆ", text)
+    text = re.sub("[Gg]", "ḡ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[İi]", "ǐ", text)
+    text = re.sub("[Uu]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Jj]", "ʒ", text)
+    text = re.sub("[Yy]", "j", text)
+    text = re.sub("[Bb]", "b", text)
+    text = re.sub("[Dd]", "d", text)
+    text = re.sub("[Ff]", "f", text)
+    text = re.sub("[Ğğ]", "ɣ", text)
+    text = re.sub("[Hh]", "h", text)
+    text = re.sub("[Xx]", "x", text)
+    text = re.sub("[Kk]", "k", text)
+    text = re.sub("[Qq]", "g", text)
+    text = re.sub("[Ll]", "l", text)
+    text = re.sub("[Mm]", "m", text)
+    text = re.sub("[Nn]", "n", text)
+    text = re.sub("[Pp]", "p", text)
+    text = re.sub("[Rr]", "r", text)
+    text = re.sub("[Ss]", "s", text)
+    text = re.sub("[Şş]", "ʃ", text)
+    text = re.sub("[Tt]", "t", text)
+    text = re.sub("[Vv]", "v", text)
+    text = re.sub("[Zz]", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("[Aa]", "ɑ", text)
+    text = re.sub("[Ee]", "e", text)
+    text = re.sub("[Əə]", "æ", text)
+    text = re.sub("[Iı]", "ɤ", text)
+    text = re.sub("[Oo]", "ɔ", text)
+    text = re.sub("[Öö]", "ɵ", text)
+    text = re.sub("[Üü]", "ʏ", text)
+
+    '''
+    rule 1:
+    if [æ], [e], [ɵ], [ʏ], [i] are followed by [l] and [l] is NOT followed by [æ], [e], [ɵ], [ʏ], or [i],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æeɵʏǐ])(l)([^æeɵʏǐ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letter "u" [ʊw] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgḡɣdʒzklłmnprstfhxʃʆǮ])", r"w\1", text)
+
+    '''
+    rule 3:
+    if the letter "i" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgḡɣdʒzklłmnprstfhxʃʆǮ])", r"i\1", text)
+
+    return text
+
+def ipa_to_azerbaijani(text):
+    # two-sound convenience consonants:
+    text = re.sub("Ǯ", "c", text)
+    text = re.sub("ʆ", "ç", text)
+    text = re.sub("g", "q", text)  # precedence issue
+    text = re.sub("ḡ", "g", text)
+
+    # single-sound consonants:
+    text = re.sub("j", "y", text)
+    text = re.sub("ʒ", "j", text)
+    text = re.sub("b", "b", text)
+    text = re.sub("d", "d", text)
+    text = re.sub("f", "f", text)
+    text = re.sub("ɣ", "ğ", text)
+    text = re.sub("h", "h", text)
+    text = re.sub("x", "x", text)
+    text = re.sub("k", "k", text)
+    text = re.sub("l", "l", text)
+    text = re.sub("m", "m", text)
+    text = re.sub("n", "n", text)
+    text = re.sub("p", "p", text)
+    text = re.sub("r", "r", text)
+    text = re.sub("s", "s", text)
+    text = re.sub("ʃ", "ş", text)
+    text = re.sub("t", "t", text)
+    text = re.sub("v", "v", text)
+    text = re.sub("z", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "a", text)
+    text = re.sub("e", "e", text)
+    text = re.sub("æ", "ə", text)
+    text = re.sub("ɤ", "ı", text)
+    text = re.sub("ǐ", "i", text)
+    text = re.sub("ɔ", "o", text)
+    text = re.sub("ɵ", "ö", text)
+    text = re.sub("ʏ", "ü", text)
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([əeöüiě])(ł)([^əeöüiě])", r"\1l\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"w([bvgğdjzkqlmnprstfhxşçc])", r"u\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"i([bcvgğdjzkqlmnprstfhxşç])", r"i\1", text)
+
+    return text
+
+# testing azerbaijani scripts
+
+def test_azerbaijani(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_azerbaijani(azerbaijani_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# turkmen scripts
+
+def turkmen_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # two-sound convenience consonants:
+    text = re.sub("[Çç]", "ʆ", text)
+    text = re.sub("[Jj]", "Ǯ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[İi]", "ǐ", text)
+    text = re.sub("[Uu]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Bb]", "b", text)
+    text = re.sub("[Dd]", "d", text)
+    text = re.sub("[Ff]", "f", text)
+    text = re.sub("[Gg]", "g", text)
+    text = re.sub("[Hh]", "h", text)
+    text = re.sub("[Žž]", "ʒ", text)
+    text = re.sub("[Kk]", "k", text)
+    text = re.sub("[Ll]", "l", text)
+    text = re.sub("[Mm]", "m", text)
+    text = re.sub("[Nn]", "n", text)
+    text = re.sub("[Ňň]", "ŋ", text)
+    text = re.sub("[Pp]", "p", text)
+    text = re.sub("[Rr]", "r", text)
+    text = re.sub("[Ss]", "s", text)  # θ
+    text = re.sub("[Şş]", "ʃ", text)
+    text = re.sub("[Tt]", "t", text)
+    text = re.sub("[Ww]", "v", text)
+    text = re.sub("[Ýý]", "j", text)
+    text = re.sub("[Zz]", "z", text)  # ð
+
+    # single-sound vowels:
+    text = re.sub("[Aa]", "ɑ", text)
+    text = re.sub("[Ää]", "æ", text)
+    text = re.sub("[Ee]", "e", text)
+    text = re.sub("[Oo]", "ɔ", text)
+    text = re.sub("[Öö]", "ɵ", text)
+    text = re.sub("[Üü]", "ʏ", text)
+    text = re.sub("[Yy]", "ɤ", text)
+
+    # rules:
+
+    '''
+    rule 1:
+    if [æ], [e], [ɵ], [ʏ], [i] are followed by [l] and [l] is NOT followed by [æ], [e], [ɵ], [ʏ], or [i],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æeɵʏǐ])(l)([^æeɵʏǐ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letter "u" [ʊw] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣqdʒzkqlłmnprstfhʃʆǮw])", r"w\1", text)
+
+    '''
+    rule 3:
+    if the letter "i" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣqdʒzkqlłmnprstfhʃʆǮ])", r"i\1", text)
+
+    # rules 4-5 are specific to Turkmen:
+
+    '''
+    rule 4:
+    a, o, u, y + k + a, o, u, y:
+    '''
+    text = re.sub(r"k([ɑɔuɤ])", r"q\1", text)
+    text = re.sub(r"([ɑɔwɤ])k", r"\1q", text)
+
+    '''
+    rule 5:
+    a, o, u, y + g + a, o, u, y:
+    '''
+    text = re.sub(r"g([ɑɔuɤ])", r"ɣ\1", text)
+    text = re.sub(r"([ɑɔwɤ])g", r"\1ɣ", text)
+
+    return text
+
+def ipa_to_turkmen(text):
+    # two-sound convenience consonants:
+    text = re.sub("j", "ý", text)  # precedence issue
+    text = re.sub("Ǯ", "j", text)
+    text = re.sub("ʆ", "ç", text)
+
+    # single-sound consonants: # w --> v can be found where the letter u anti-rule is
+    text = re.sub("b", "b", text)
+    text = re.sub("d", "d", text)
+    text = re.sub("f", "f", text)
+    text = re.sub("g", "g", text)
+    text = re.sub("ɣ", "g", text)
+    text = re.sub("h", "h", text)
+    text = re.sub("ʒ", "ž", text)
+    text = re.sub("k", "k", text)
+    text = re.sub("q", "k", text)
+    text = re.sub("l", "l", text)
+    text = re.sub("m", "m", text)
+    text = re.sub("n", "n", text)
+    text = re.sub("ŋ", "ň", text)
+    text = re.sub("p", "p", text)
+    text = re.sub("r", "r", text)
+    text = re.sub("s", "s", text)
+    text = re.sub("ʃ", "ş", text)
+    text = re.sub("t", "t", text)
+    text = re.sub("z", "z", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "a", text)
+    text = re.sub("e", "e", text)
+    text = re.sub("æ", "ä", text)
+    text = re.sub("ǐ", "i", text)
+    text = re.sub("ɔ", "o", text)
+    text = re.sub("ɵ", "ö", text)
+    text = re.sub("ʏ", "ü", text)
+    text = re.sub("ɤ", "y", text)
+
+    # anti-rules:
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([äeöüiě])(ł)([^äeöüiě])", r"\1l\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"w([bdfghžklmnňprsştýzjçɣqv])", r"u\1", text)  # precedence issue
+    text = re.sub("v", "w", text)  # precedence issue
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"i([bdfghžklmnňprsştwýzjçɣq])", r"i\1", text)
+
+    return text
+
+# testing turkmen scripts
+
+def test_turkmen(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_turkmen(turkmen_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# tatar scripts
+
+def tatar_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # three-sound convenience vowels:
+    text = re.sub("[Юю]", "ǔ", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("[Цц]", "š", text)
+    text = re.sub("[Чч]", "ʆ", text)
+    text = re.sub("[Җҗ]", "Ǯ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Яя]", "ǎ", text)
+    text = re.sub("[Ее]", "ě", text)
+    text = re.sub("[Ёё]", "ǒ", text)
+    text = re.sub("[Ии]", "ǐ", text)
+    text = re.sub("[Уу]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Бб]", "b", text)
+    text = re.sub("[Вв]", "v", text)
+    text = re.sub("[Гг]", "g", text)
+    text = re.sub("[Дд]", "d", text)
+    text = re.sub("[Жж]", "ʒ", text)
+    text = re.sub("[Зз]", "z", text)
+    text = re.sub("[Йй]", "j", text)
+    text = re.sub("[Кк]", "k", text)
+    text = re.sub("[Лл]", "l", text)
+    text = re.sub("[Мм]", "m", text)
+    text = re.sub("[Нн]", "n", text)
+    text = re.sub("[Ңң]", "ŋ", text)
+    text = re.sub("[Пп]", "p", text)
+    text = re.sub("[Рр]", "r", text)
+    text = re.sub("[Сс]", "s", text)
+    text = re.sub("[Тт]", "t", text)
+    text = re.sub("[Фф]", "f", text)
+    text = re.sub("[Хх]", "x", text)
+    text = re.sub("[Һһ]", "h", text)
+    text = re.sub("[Шш]", "ʃ", text)
+    text = re.sub("[Щщ]", "ɕ", text)
+    text = re.sub("[Ъъ]", "ʔ", text)
+    text = re.sub("[Ьь]", "ʲ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Аа]", "ɑ", text)
+    text = re.sub("[Әә]", "æ", text)
+    text = re.sub("[Оо]", "ɔ", text)
+    text = re.sub("[Өө]", "ɵ", text)
+    text = re.sub("[Үү]", "ʏ", text)
+    text = re.sub("[Ыы]", "ɤ", text)
+    text = re.sub("[Ээ]", "e", text)
+
+    # rules 1-4 are similar to those for Kazakh:
+
+    '''
+    rule 1:
+    if [ě], [ɵ], [ʏ], are followed by [l] and [l] is NOT followed by [ě], [ɵ], [ʏ], or [ʲ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æɵʏě])(l)([^æɵʏěʲ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letters "о" and "ө", [ɔ] and [ɵ] at the beginning of a word are followed by [w] (e.g., осы [wɔsɤ] not [ɔsɤ], өзі [wɵzɪ] not [ɵzɪ]).
+    '''
+    text = re.sub(r"\b([ɔɵ])", r"w\1", text)
+
+    '''
+    rule 3
+    if the letter "у" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆǮʲ])", r"w\1", text)
+
+    '''
+    rule 4:
+    if the letter "и" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆǮʲ])", r"i\1", text)
+
+    # rules 5-6 are specific to Tatar:
+
+    '''
+    rule 5:
+    а, о, у, ы, ъ + к + а, о, у, ы, ъ
+    '''
+    text = re.sub(r"k([ɑɔwɤʔ])", r"q\1", text)
+    text = re.sub(r"([ɑɔwɤʔ])k", r"\1q", text)
+
+    '''
+    rule 6:
+    а, о, у, ы, ъ + г + а, о, у, ы, ъ
+    '''
+    text = re.sub(r"g([ɑɔwɤʔ])", r"ɣ\1", text)
+    text = re.sub(r"([ɑɔwɤʔ])g", r"\1ɣ", text)
+
+    return text
+
+def ipa_to_tatar(text):
+    # three-sound convenience vowels:
+    text = re.sub("ǔ", "ю", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("š", "ц", text)
+    text = re.sub("ʆ", "ч", text)
+    text = re.sub("Ǯ", "җ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ǎ", "я", text)
+    text = re.sub("ě", "е", text)
+    text = re.sub("ǒ", "ё", text)
+    text = re.sub("ǐ", "и", text)
+    text = re.sub("u", "у", text)
+
+    # single-sound consonants:
+    text = re.sub("b", "б", text)
+    text = re.sub("v", "в", text)
+    text = re.sub("g", "г", text)
+    text = re.sub("ɣ", "г", text)
+    text = re.sub("d", "д", text)
+    text = re.sub("ʒ", "ж", text)
+    text = re.sub("z", "з", text)
+    text = re.sub("j", "й", text)
+    text = re.sub("k", "к", text)
+    text = re.sub("l", "л", text)
+    text = re.sub("m", "м", text)
+    text = re.sub("n", "н", text)
+    text = re.sub("ŋ", "ң", text)
+    text = re.sub("p", "п", text)
+    text = re.sub("q", "к", text)
+    text = re.sub("r", "р", text)
+    text = re.sub("s", "с", text)
+    text = re.sub("t", "т", text)
+    text = re.sub("f", "ф", text)
+    text = re.sub("x", "х", text)
+    text = re.sub("h", "һ", text)
+    text = re.sub("ʃ", "ш", text)
+    text = re.sub("ɕ", "щ", text)
+    text = re.sub("ʔ", "ъ", text)
+    text = re.sub("ʲ", "ь", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "а", text)
+    text = re.sub("æ", "ә", text)
+    text = re.sub("ɔ", "о", text)
+    text = re.sub("ɵ", "ө", text)
+    text = re.sub("ʏ", "ү", text)
+    text = re.sub("ɤ", "ы", text)
+    text = re.sub("e", "э", text)
+
+    # anti-rules 1-4 are similar to those for Kazakh:
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([әөүе])(ł)([^әөүеʲ])", r"\1л\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"\bw([оө])", r"\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"w([бвгдзйклмнңпрстфхһцчшщъьчцжҗqɣ])", r"у\1", text)
+
+    '''
+    anti-rule 4:
+    '''
+    text = re.sub(r"i([бвгдзйклмнңпрстфхһцчшщъьчцжҗqɣ])", r"и\1", text)
+
+    return text
+
+# testing tatar scripts
+
+def test_tatar(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_tatar(tatar_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# bashkir scripts
+
+def bashkir_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we shall later convert them to conventional symbols.
+
+    # three-sound convenience vowels:
+    text = re.sub("[Юю]", "ǔ", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("[Цц]", "š", text)
+    text = re.sub("[Чч]", "ʆ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Яя]", "ǎ", text)
+    text = re.sub("[Ее]", "ě", text)
+    text = re.sub("[Ёё]", "ǒ", text)
+    text = re.sub("[Ии]", "ǐ", text)
+    text = re.sub("[Уу]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Бб]", "b", text)
+    text = re.sub("[Вв]", "v", text)
+    text = re.sub("[Гг]", "g", text)
+    text = re.sub("[Ғғ]", "ɣ", text)
+    text = re.sub("[Дд]", "d", text)
+    text = re.sub("[Ҙҙ]", "z", text)
+    text = re.sub("[Жж]", "ʒ", text)
+    text = re.sub("[Зз]", "z", text)
+    text = re.sub("[Йй]", "j", text)
+    text = re.sub("[Кк]", "k", text)
+    text = re.sub("[Ҡҡ]", "q", text)
+    text = re.sub("[Лл]", "l", text)
+    text = re.sub("[Мм]", "m", text)
+    text = re.sub("[Нн]", "n", text)
+    text = re.sub("[Ңң]", "ŋ", text)
+    text = re.sub("[Пп]", "p", text)
+    text = re.sub("[Рр]", "r", text)
+    text = re.sub("[Сс]", "s", text)
+    text = re.sub("[Ҫҫ]", "s", text)
+    text = re.sub("[Тт]", "t", text)
+    text = re.sub("[Хх]", "x", text)
+    text = re.sub("[Фф]", "f", text)
+    text = re.sub("[Һһ]", "h", text)
+    text = re.sub("[Шш]", "ʃ", text)
+    text = re.sub("[Щщ]", "ɕ", text)
+    text = re.sub("[Ъъ]", "ʔ", text)
+    text = re.sub("[Ьь]", "ʲ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Аа]", "ɑ", text)
+    text = re.sub("[Әә]", "æ", text)
+    text = re.sub("[Оо]", "ɔ", text)
+    text = re.sub("[Өө]", "ɵ", text)
+    text = re.sub("[Үү]", "ʏ", text)
+    text = re.sub("[Ыы]", "ɤ", text)
+    text = re.sub("[Ээ]", "e", text)
+
+    # rules 1-4 are similar to those for Kazakh:
+
+    '''
+    rule 1:
+    if [ě], [ɵ], [ʏ], are followed by [l] and [l] is NOT followed by [ě], [ɵ], [ʏ], or [ʲ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æɵʏě])(l)([^æɵʏěʲ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letters "о" and "ө", [ɔ] and [ɵ] at the beginning of a word are followed by [w] (e.g., осы [wɔsɤ] not [ɔsɤ], өзі [wɵzɪ] not [ɵzɪ]).
+    '''
+    text = re.sub(r"\b([ɔɵ])", r"w\1", text)
+
+    '''
+    rule 3
+    if the letter "у" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆʲ])", r"w\1", text)
+
+    '''
+    rule 4:
+    if the letter "и" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdʒzjkqlłmnŋprstfxhʃɕʔšʆʲ])", r"i\1", text)
+
+    return text
+
+def ipa_to_bashkir(text):
+    # three-sound convenience vowels:
+    text = re.sub("ǔ", "ю", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("š", "ц", text)
+    text = re.sub("ʆ", "ч", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ě", "е", text)
+    text = re.sub("ǒ", "ё", text)
+    text = re.sub("ǐ", "и", text)
+    text = re.sub("u", "у", text)
+    text = re.sub("ǎ", "я", text)
+
+    # single-sound consonants:
+    text = re.sub("b", "б", text)
+    text = re.sub("v", "в", text)
+    text = re.sub("g", "г", text)
+    text = re.sub("ɣ", "ғ", text)
+    text = re.sub("d", "д", text)
+    text = re.sub("z", "з", text)
+    text = re.sub("ʒ", "ж", text)
+    text = re.sub("j", "й", text)
+    text = re.sub("k", "к", text)
+    text = re.sub("q", "ҡ", text)
+    text = re.sub("l", "л", text)
+    text = re.sub("m", "м", text)
+    text = re.sub("n", "н", text)
+    text = re.sub("ŋ", "ң", text)
+    text = re.sub("p", "п", text)
+    text = re.sub("r", "р", text)
+    text = re.sub("s", "с", text)
+    text = re.sub("t", "т", text)
+    text = re.sub("f", "ф", text)
+    text = re.sub("x", "х", text)
+    text = re.sub("h", "һ", text)
+    text = re.sub("ʃ", "ш", text)
+    text = re.sub("ɕ", "щ", text)
+    text = re.sub("ʔ", "ъ", text)
+    text = re.sub("ʲ", "ь", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "а", text)
+    text = re.sub("æ", "ә", text)
+    text = re.sub("ɔ", "о", text)
+    text = re.sub("ɵ", "ө", text)
+    text = re.sub("ʏ", "ү", text)
+    text = re.sub("ɤ", "ы", text)
+    text = re.sub("e", "э", text)
+
+    # anti-rules 1-4 are similar to those for Kazakh:
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([әөүе])(ł)([^әөүеʲ])", r"\1л\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"\bw([оө])", r"\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"w([бвгғдзйкҡлмнңпрстфхһцчшщъьчцж])", r"у\1", text)
+
+    '''
+    anti-rule 4:
+
+
+    '''
+    text = re.sub(r"i([бвгғдзйкҡлмнңпрстфхһцчшщъьчцж])", r"и\1", text)
+
+    return text
+
+# testing bashkir scripts
+
+def test_bashkir(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_bashkir(bashkir_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# sakha scripts
+
+def sakha_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # three-sound convenience vowels:
+    text = re.sub("[Юю]", "ǔ", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("[Цц]", "š", text)
+    text = re.sub("[Чч]", "ʆ", text)
+    text = re.sub("ДЬ", "Ǯ", text)
+    text = re.sub("дь", "Ǯ", text)
+    text = re.sub("Дь", "Ǯ", text)
+    text = re.sub("дЬ", "Ǯ", text)
+    text = re.sub("НЬ", "ɲ", text)
+    text = re.sub("нь", "ɲ", text)
+    text = re.sub("Нь", "ɲ", text)
+    text = re.sub("нЬ", "ɲ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Яя]", "ǎ", text)
+    text = re.sub("[Ее]", "ě", text)
+    text = re.sub("[Ёё]", "ǒ", text)
+    text = re.sub("[Ии]", "ǐ", text)
+    text = re.sub("[Уу]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Бб]", "b", text)
+    text = re.sub("[Вв]", "v", text)
+    text = re.sub("[Гг]", "g", text)
+    text = re.sub("[Ҕҕ]", "ɣ", text)
+    text = re.sub("[Дд]", "d", text)
+    text = re.sub("[Жж]", "ʒ", text)
+    text = re.sub("[Зз]", "z", text)
+    text = re.sub("[Йй]", "j", text)
+    text = re.sub("[Кк]", "k", text)
+    text = re.sub("[Лл]", "l", text)
+    text = re.sub("[Мм]", "m", text)
+    text = re.sub("[Нн]", "n", text)
+    text = re.sub("[Ҥҥ]", "ŋ", text)
+    text = re.sub("[Пп]", "p", text)
+    text = re.sub("[Рр]", "r", text)
+    text = re.sub("[Сс]", "s", text)
+    text = re.sub("[Тт]", "t", text)
+    text = re.sub("[Хх]", "x", text)
+    text = re.sub("[Фф]", "f", text)
+    text = re.sub("[Һһ]", "h", text)
+    text = re.sub("[Шш]", "ʃ", text)
+    text = re.sub("[Щщ]", "ɕ", text)
+    text = re.sub("[Ъъ]", "ʔ", text)
+    text = re.sub("[Ьь]", "ʲ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Аа]", "ɑ", text)
+    text = re.sub("[Әә]", "æ", text)
+    text = re.sub("[Оо]", "ɔ", text)
+    text = re.sub("[Өө]", "ɵ", text)
+    text = re.sub("[Үү]", "ʏ", text)
+    text = re.sub("[Ыы]", "ɤ", text)
+    text = re.sub("[Ээ]", "e", text)
+
+    # rules 1-4 are similar to those for Kazakh:
+
+    '''
+    rule 1:
+    if [ě], [ɵ], [ʏ], are followed by [l] and [l] is NOT followed by [ě], [ɵ], [ʏ], or [ʲ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æɵʏě])(l)([^æɵʏěʲ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letters "о" and "ө", [ɔ] and [ɵ] at the beginning of a word are followed by [w] (e.g., осы [wɔsɤ] not [ɔsɤ], өзі [wɵzɪ] not [ɵzɪ]).
+    '''
+    text = re.sub(r"\b([ɔɵ])", r"w\1", text)
+
+    '''
+    rule 3
+    if the letter "у" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdʒzjklłmnŋɲprstfxhʃɕʔšʆǮʲ])", r"w\1", text)
+
+    '''
+    rule 4:
+    if the letter "и" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdʒzjklłmnŋɲprstfxhʃɕʔšʆǮʲ])", r"i\1", text)
+
+    return text
+
+def ipa_to_sakha(text):
+    # three-sound convenience vowels:
+    text = re.sub("ǔ", "ю", text)
+
+    # two-sound convenience consonants:
+    text = re.sub("š", "ц", text)
+    text = re.sub("ʆ", "ч", text)
+    text = re.sub("Ǯ", "дь", text)
+    text = re.sub("ɲ", "нь", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ě", "е", text)
+    text = re.sub("ǒ", "ё", text)
+    text = re.sub("ǐ", "и", text)
+    text = re.sub("u", "у", text)
+    text = re.sub("ǎ", "я", text)
+
+    # single-sound consonants:
+    text = re.sub("b", "б", text)
+    text = re.sub("v", "в", text)
+    text = re.sub("g", "г", text)
+    text = re.sub("ɣ", "ҕ", text)
+    text = re.sub("d", "д", text)
+    text = re.sub("z", "з", text)
+    text = re.sub("ʒ", "ж", text)
+    text = re.sub("j", "й", text)
+    text = re.sub("k", "к", text)
+    text = re.sub("l", "л", text)
+    text = re.sub("m", "м", text)
+    text = re.sub("n", "н", text)
+    text = re.sub("ŋ", "ҥ", text)
+    text = re.sub("p", "п", text)
+    text = re.sub("r", "р", text)
+    text = re.sub("s", "с", text)
+    text = re.sub("t", "т", text)
+    text = re.sub("f", "ф", text)
+    text = re.sub("x", "х", text)
+    text = re.sub("h", "һ", text)
+    text = re.sub("ʃ", "ш", text)
+    text = re.sub("ɕ", "щ", text)
+    text = re.sub("ʔ", "ъ", text)
+    text = re.sub("ʲ", "ь", text)
+
+    # single-sound vowels:
+    text = re.sub("ɑ", "а", text)
+    text = re.sub("æ", "ә", text)
+    text = re.sub("ɔ", "о", text)
+    text = re.sub("ɵ", "ө", text)
+    text = re.sub("ʏ", "ү", text)
+    text = re.sub("ɤ", "ы", text)
+    text = re.sub("e", "э", text)
+
+    # anti-rules 1-4 are similar to those for Kazakh:
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([әөүе])(ł)([^әөүеʲ])", r"\1л\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"\bw([оө])", r"\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"w(дь)", r"у\1", text)
+    text = re.sub(r"w(нь)", r"у\1", text)
+    text = re.sub(r"w([бвгҕдзйклмнҥпрстфхһцчшщъьчцж])", r"у\1", text)
+
+    '''
+    anti-rule 4:
+    '''
+    text = re.sub(r"i(дь)", r"и\1", text)
+    text = re.sub(r"i(нь)", r"и\1", text)
+    text = re.sub(r"i([бвгҕдзйклмнҥпрстфхһцчшщъьчцж])", r"и\1", text)
+
+    return text
+
+# testing sakha scripts
+
+# testing bashkir scripts
+
+def test_sakha(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_sakha(sakha_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)
+
+# uyghur scripts
+
+def uyghur_to_ipa(text):
+    # we shall begin with sound combinations:
+    # the longer a combination, the upper it is on the list
+    # single sounds should go to the bottom, with consonants taking precedence over vowels.
+    # consonants are less likely to change than vowels.
+
+    # for convenience, we shall use single symbols to denote multiple-sound combinations.
+    # we can later convert them to conventional symbols.
+
+    # two-sound convenience consonants:
+    text = re.sub("[Jj]", "Ǯ", text)
+    text = re.sub("Ch", "ʆ", text)
+    text = re.sub("ch", "ʆ", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("[Ii]", "ǐ", text)
+    text = re.sub("[Uu]", "u", text)
+
+    # single-sound consonants:
+    text = re.sub("[Bb]", "b", text)
+    text = re.sub("[Dd]", "d", text)
+    text = re.sub("[Ff]", "f", text)
+    text = re.sub("Gh", "ɣ", text)
+    text = re.sub("gh", "ɣ", text)
+    text = re.sub("[Gg]", "g", text)
+    text = re.sub("[Hh]", "h", text)
+    text = re.sub("[Kk]", "k", text)
+    text = re.sub("[Ll]", "l", text)
+    text = re.sub("[Mm]", "m", text)
+    text = re.sub("[Nn]", "n", text)
+    text = re.sub("Ng", "ŋ", text)
+    text = re.sub("ng", "ŋ", text)
+    text = re.sub("[Pp]", "p", text)
+    text = re.sub("[Qq]", "q", text)
+    text = re.sub("[Rr]", "r", text)
+    text = re.sub("[Ss]", "s", text)
+    text = re.sub("Sh", "ʃ", text)
+    text = re.sub("sh", "ʃ", text)
+    text = re.sub("[Tt]", "t", text)
+    text = re.sub("[Ww]", "v", text)
+    text = re.sub("[Xx]", "x", text)
+    text = re.sub("[Yy]", "j", text)
+    text = re.sub("[Zz]", "z", text)
+    text = re.sub("Zh", "ʒ", text)
+    text = re.sub("zh", "ʒ", text)
+
+    # single-sound vowels:
+    text = re.sub("[Aa]", "ɑ", text)
+    text = re.sub("[Ee]", "æ", text)
+    text = re.sub("[ËÉëé]", "e", text)
+    text = re.sub("[Oo]", "ɔ", text)
+    text = re.sub("[Öö]", "ɵ", text)
+    text = re.sub("[Üü]", "ʏ", text)
+
+    # hard sign
+    text = re.sub("'", "ʔ", text)
+
+    '''
+    rule 1:
+    if [æ], [e], [ɵ], [ǐ] are followed by [l] and [l] is NOT followed by [æ], [e], [ɵ], [ǐ],
+    use [ł] instead of [l].
+    '''
+    text = re.sub(r"([æɵǐeʏ])(l)([^æɵǐeʏ])", r"\1ł\3", text)
+
+    '''
+    rule 2:
+    if the letter "u" [u] is followed by consonants, use [w] instead of [u].
+    '''
+    text = re.sub(r"u([bvgɣdzjkqlłmnŋprstfxhʃʆǮʒ])", r"w\1", text)
+
+    '''
+    rule 3:
+    if the letter "i" [ǐ] is followed by consonants, use [i] instead of [ǐ].
+    '''
+    text = re.sub(r"ǐ([bvgɣdzjkqlłmnŋprstfxhʃʆǮʒ])", r"i\1", text)
+
+    return text
+
+def ipa_to_uyghur(text):
+    # two-sound convenience consonants:
+    text = re.sub("j", "y", text)  # exception! precedence issue
+    text = re.sub("Ǯ", "j", text)
+    text = re.sub("ʆ", "ch", text)
+    text = re.sub("ʒ", "zh", text)
+
+    # two-sound convenience vowels:
+    text = re.sub("ǐ", "i", text)
+    text = re.sub("u", "u", text)
+
+    # single-sound convenience consonants:
+    text = re.sub("b", "b", text)
+    text = re.sub("d", "d", text)
+    text = re.sub("f", "f", text)
+    text = re.sub("g", "g", text)
+    text = re.sub("ɣ", "gh", text)
+    text = re.sub("h", "h", text)
+    text = re.sub("k", "k", text)
+    text = re.sub("l", "l", text)
+    text = re.sub("m", "m", text)
+    text = re.sub("n", "n", text)
+    text = re.sub("ŋ", "ng", text)
+    text = re.sub("p", "p", text)
+    text = re.sub("q", "q", text)
+    text = re.sub("r", "r", text)
+    text = re.sub("s", "s", text)
+    text = re.sub("ʃ", "sh", text)
+    text = re.sub("t", "t", text)
+    text = re.sub("v", "w", text)
+    text = re.sub("x", "x", text)
+    text = re.sub("z", "z", text)
+
+    # single-sound convenience vowels:
+    text = re.sub("ɑ", "a", text)
+    text = re.sub("e", "ë", text) # precedence
+    text = re.sub("æ", "e", text)
+    text = re.sub("ɵ", "ö", text)
+    text = re.sub("ɔ", "o", text)
+    text = re.sub("ʏ", "ü", text)
+
+    # hard sign
+    text = re.sub("ʔ", "'", text)
+
+    '''
+    anti-rule 1:
+    '''
+    text = re.sub(r"([eëiöü])(ł)([^eëiöü])", r"\1l\3", text)
+
+    '''
+    anti-rule 2:
+    '''
+    text = re.sub(r"w([bcvgdjzklmnpqrstfhyx])", r"u\1", text)
+
+    '''
+    anti-rule 3:
+    '''
+    text = re.sub(r"i([bcvgdjzklmnpqrstfhyx])", r"i\1", text)
+
+    return text
+
+# testing uyghur scripts
+
+def test_uyghur(text):
+    input_text = text.lower().split()
+    output_text = ipa_to_uyghur(uyghur_to_ipa(text)).split()
+
+    input_difference = []
+    output_difference = []
+    for item in input_text:
+        if item not in output_text:
+            input_difference.append(item)
+    for item in output_text:
+        if item not in input_text:
+            output_difference.append(item)
+
+    if input_text == output_text:
+        print("input text and output text -- identical")
+    else:
+        print("input text and output text -- different")
+        print("input:", input_difference)
+        print("output:", output_difference)

turkicTTS_utils.py

@@ -0,0 +1,24 @@
+import turkicTTS_ipa_convert as ipa_convert
+
+dispatcher = {'kazakh' : ipa_convert.kazakh_to_ipa, 
+              'turkish' : ipa_convert.turkish_to_ipa,
+              'azerbaijani': ipa_convert.azerbaijani_to_ipa,
+              'kyrgyz' : ipa_convert.kyrgyz_to_ipa,              
+              'uzbek' : ipa_convert.uzbek_to_ipa,
+              'turkmen' : ipa_convert.turkmen_to_ipa,
+              'tatar' : ipa_convert.tatar_to_ipa,
+              'bashkir' : ipa_convert.bashkir_to_ipa,
+              'sakha' : ipa_convert.sakha_to_ipa,
+              'uyghur' : ipa_convert.uyghur_to_ipa,              
+              'tts_sent' : ipa_convert.ipa_to_kazakh}
+
+def call_func(x, func):
+    try:
+        return dispatcher[func](x)
+    except:
+        return "Invalid function"
+
+def normalization(x, lang="kazakh"):
+    ipa_text = call_func(x, lang)
+    kz_text = call_func(ipa_text, 'tts_sent')
+    return kz_text