Merge master into main

.gitattributes

@@ -0,0 +1,31 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,6 +1,6 @@
 ---
 title: Audiobook Gen
-emoji: 📚 
+emoji: 📚
 colorFrom: blue
 colorTo: indigo
 sdk: streamlit

app.py

@@ -41,7 +41,7 @@ if st.button('Click to run!'):
         text, file_title = file_readers.read_epub(uploaded_file)
     elif file_ext == 'text/plain':
         file = uploaded_file.read()
-        text = [file_readers.preprocess_text(file)]
+        text = file_readers.preprocess_text(file)
     elif file_ext == 'application/pdf':
         text = file_readers.read_pdf(uploaded_file)
     else:
@@ -64,3 +64,11 @@ if st.button('Click to run!'):
             file_name=title_name,
             mime="application/zip"
         )
+
+    with open(zip_file, "rb") as fp:
+        btn = st.download_button(
+            label="Download Audiobook",
+            data=fp,
+            file_name=title_name,
+            mime="application/zip"
+        )

latest_silero_models.yml

@@ -0,0 +1,563 @@
+# pre-trained STT models
+stt_models:
+  en:
+    latest:
+      meta:
+        name: "en_v6"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v6.jit"
+      onnx: "https://models.silero.ai/models/en/en_v5.onnx"
+      jit_q: "https://models.silero.ai/models/en/en_v6_q.jit"
+      jit_xlarge: "https://models.silero.ai/models/en/en_v6_xlarge.jit"
+      onnx_xlarge: "https://models.silero.ai/models/en/en_v6_xlarge.onnx"
+    v6:
+      meta:
+        name: "en_v6"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v6.jit"
+      onnx: "https://models.silero.ai/models/en/en_v5.onnx"
+      jit_q: "https://models.silero.ai/models/en/en_v6_q.jit"
+      jit_xlarge: "https://models.silero.ai/models/en/en_v6_xlarge.jit"
+      onnx_xlarge: "https://models.silero.ai/models/en/en_v6_xlarge.onnx"
+    v5:
+      meta:
+        name: "en_v5"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v5.jit"
+      onnx: "https://models.silero.ai/models/en/en_v5.onnx"
+      onnx_q: "https://models.silero.ai/models/en/en_v5_q.onnx"
+      jit_q: "https://models.silero.ai/models/en/en_v5_q.jit"
+      jit_xlarge: "https://models.silero.ai/models/en/en_v5_xlarge.jit"
+      onnx_xlarge: "https://models.silero.ai/models/en/en_v5_xlarge.onnx"
+    v4_0:
+      meta:
+        name: "en_v4_0"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit_large: "https://models.silero.ai/models/en/en_v4_0_jit_large.model"
+      onnx_large: "https://models.silero.ai/models/en/en_v4_0_large.onnx"
+    v3:
+      meta:
+        name: "en_v3"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v3_jit.model"
+      onnx: "https://models.silero.ai/models/en/en_v3.onnx"
+      jit_q: "https://models.silero.ai/models/en/en_v3_jit_q.model"
+      jit_skip: "https://models.silero.ai/models/en/en_v3_jit_skips.model"
+      jit_large: "https://models.silero.ai/models/en/en_v3_jit_large.model"
+      onnx_large: "https://models.silero.ai/models/en/en_v3_large.onnx"
+      jit_xsmall: "https://models.silero.ai/models/en/en_v3_jit_xsmall.model"
+      jit_q_xsmall: "https://models.silero.ai/models/en/en_v3_jit_q_xsmall.model"
+      onnx_xsmall: "https://models.silero.ai/models/en/en_v3_xsmall.onnx"
+    v2:
+      meta:
+        name: "en_v2"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v2_jit.model"
+      onnx: "https://models.silero.ai/models/en/en_v2.onnx"
+      tf: "https://models.silero.ai/models/en/en_v2_tf.tar.gz"
+    v1:
+      meta:
+        name: "en_v1"
+        sample: "https://models.silero.ai/examples/en_sample.wav"
+      labels: "https://models.silero.ai/models/en/en_v1_labels.json"
+      jit: "https://models.silero.ai/models/en/en_v1_jit.model"
+      onnx: "https://models.silero.ai/models/en/en_v1.onnx"
+      tf: "https://models.silero.ai/models/en/en_v1_tf.tar.gz"
+  de:
+    latest:
+      meta:
+        name: "de_v1"
+        sample: "https://models.silero.ai/examples/de_sample.wav"
+      labels: "https://models.silero.ai/models/de/de_v1_labels.json"
+      jit: "https://models.silero.ai/models/de/de_v1_jit.model"
+      onnx: "https://models.silero.ai/models/de/de_v1.onnx"
+      tf: "https://models.silero.ai/models/de/de_v1_tf.tar.gz"
+    v1:
+      meta:
+        name: "de_v1"
+        sample: "https://models.silero.ai/examples/de_sample.wav"
+      labels: "https://models.silero.ai/models/de/de_v1_labels.json"
+      jit_large: "https://models.silero.ai/models/de/de_v1_jit.model"
+      onnx: "https://models.silero.ai/models/de/de_v1.onnx"
+      tf: "https://models.silero.ai/models/de/de_v1_tf.tar.gz"
+    v3:
+      meta:
+        name: "de_v3"
+        sample: "https://models.silero.ai/examples/de_sample.wav"
+      labels: "https://models.silero.ai/models/de/de_v1_labels.json"
+      jit_large: "https://models.silero.ai/models/de/de_v3_large.jit"
+    v4:
+      meta:
+        name: "de_v4"
+        sample: "https://models.silero.ai/examples/de_sample.wav"
+      labels: "https://models.silero.ai/models/de/de_v1_labels.json"
+      jit_large: "https://models.silero.ai/models/de/de_v4_large.jit"
+      onnx_large: "https://models.silero.ai/models/de/de_v4_large.onnx"
+  es:
+    latest:
+      meta:
+        name: "es_v1"
+        sample: "https://models.silero.ai/examples/es_sample.wav"
+      labels: "https://models.silero.ai/models/es/es_v1_labels.json"
+      jit: "https://models.silero.ai/models/es/es_v1_jit.model"
+      onnx: "https://models.silero.ai/models/es/es_v1.onnx"
+      tf: "https://models.silero.ai/models/es/es_v1_tf.tar.gz"
+  ua:
+    latest:
+      meta:
+        name: "ua_v3"
+        sample: "https://models.silero.ai/examples/ua_sample.wav"
+        credits:
+          datasets:
+            speech-recognition-uk: https://github.com/egorsmkv/speech-recognition-uk
+      labels: "https://models.silero.ai/models/ua/ua_v1_labels.json"
+      jit: "https://models.silero.ai/models/ua/ua_v3_jit.model"
+      jit_q: "https://models.silero.ai/models/ua/ua_v3_jit_q.model"
+      onnx: "https://models.silero.ai/models/ua/ua_v3.onnx"
+    v3:
+      meta:
+        name: "ua_v3"
+        sample: "https://models.silero.ai/examples/ua_sample.wav"
+        credits:
+          datasets:
+            speech-recognition-uk: https://github.com/egorsmkv/speech-recognition-uk
+      labels: "https://models.silero.ai/models/ua/ua_v1_labels.json"
+      jit: "https://models.silero.ai/models/ua/ua_v3_jit.model"
+      jit_q: "https://models.silero.ai/models/ua/ua_v3_jit_q.model"
+      onnx: "https://models.silero.ai/models/ua/ua_v3.onnx"
+    v1:
+      meta:
+        name: "ua_v1"
+        sample: "https://models.silero.ai/examples/ua_sample.wav"
+        credits:
+          datasets:
+            speech-recognition-uk: https://github.com/egorsmkv/speech-recognition-uk
+      labels: "https://models.silero.ai/models/ua/ua_v1_labels.json"
+      jit: "https://models.silero.ai/models/ua/ua_v1_jit.model"
+      jit_q: "https://models.silero.ai/models/ua/ua_v1_jit_q.model"
+tts_models:
+  ru:
+    v3_1_ru:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v3_1_ru.pt'
+        sample_rate: [8000, 24000, 48000]
+    ru_v3:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/ru_v3.pt'
+        sample_rate: [8000, 24000, 48000]
+    aidar_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_aidar.pt'
+        sample_rate: [8000, 16000]
+    aidar_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_aidar_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_aidar_8000.jit'
+        sample_rate: 8000
+    aidar_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_aidar_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_aidar_16000.jit'
+        sample_rate: 16000
+    baya_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_baya.pt'
+        sample_rate: [8000, 16000]
+    baya_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_baya_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_baya_8000.jit'
+        sample_rate: 8000
+    baya_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_baya_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_baya_16000.jit'
+        sample_rate: 16000
+    irina_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_irina.pt'
+        sample_rate: [8000, 16000]
+    irina_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_irina_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_irina_8000.jit'
+        sample_rate: 8000
+    irina_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_irina_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_irina_16000.jit'
+        sample_rate: 16000
+    kseniya_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_kseniya.pt'
+        sample_rate: [8000, 16000]
+    kseniya_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_kseniya_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_kseniya_8000.jit'
+        sample_rate: 8000
+    kseniya_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_kseniya_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_kseniya_16000.jit'
+        sample_rate: 16000
+    natasha_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_natasha.pt'
+        sample_rate: [8000, 16000]
+    natasha_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_natasha_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_natasha_8000.jit'
+        sample_rate: 8000
+    natasha_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_natasha_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_natasha_16000.jit'
+        sample_rate: 16000
+    ruslan_v2:
+      latest:
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        package: 'https://models.silero.ai/models/tts/ru/v2_ruslan.pt'
+        sample_rate: [8000, 16000]
+    ruslan_8khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_ruslan_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_ruslan_8000.jit'
+        sample_rate: 8000
+    ruslan_16khz:
+      latest:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_ruslan_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~абвгдеёжзийклмнопрстуфхцчшщъыьэюя +.,!?…:;–'
+        example: 'В н+едрах т+ундры в+ыдры в г+етрах т+ырят в в+ёдра +ядра к+едров.'
+        jit: 'https://models.silero.ai/models/tts/ru/v1_ruslan_16000.jit'
+        sample_rate: 16000
+  en:
+    v3_en:
+      latest:
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        package: 'https://models.silero.ai/models/tts/en/v3_en.pt'
+        sample_rate: [8000, 24000, 48000]
+    v3_en_indic:
+      latest:
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        package: 'https://models.silero.ai/models/tts/en/v3_en_indic.pt'
+        sample_rate: [8000, 24000, 48000]
+    lj_v2:
+      latest:
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        package: 'https://models.silero.ai/models/tts/en/v2_lj.pt'
+        sample_rate: [8000, 16000]
+    lj_8khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyz .,!?…:;–'
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        jit: 'https://models.silero.ai/models/tts/en/v1_lj_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyz .,!?…:;–'
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        jit: 'https://models.silero.ai/models/tts/en/v1_lj_8000.jit'
+        sample_rate: 8000
+    lj_16khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyz .,!?…:;–'
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        jit: 'https://models.silero.ai/models/tts/en/v1_lj_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyz .,!?…:;–'
+        example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+        jit: 'https://models.silero.ai/models/tts/en/v1_lj_16000.jit'
+        sample_rate: 16000
+  de:
+    v3_de:
+      latest:
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        package: 'https://models.silero.ai/models/tts/de/v3_de.pt'
+        sample_rate: [8000, 24000, 48000]
+    thorsten_v2:
+      latest:
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        package: 'https://models.silero.ai/models/tts/de/v2_thorsten.pt'
+        sample_rate: [8000, 16000]
+    thorsten_8khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzäöüß .,!?…:;–'
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        jit: 'https://models.silero.ai/models/tts/de/v1_thorsten_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzäöüß .,!?…:;–'
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        jit: 'https://models.silero.ai/models/tts/de/v1_thorsten_8000.jit'
+        sample_rate: 8000
+    thorsten_16khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzäöüß .,!?…:;–'
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        jit: 'https://models.silero.ai/models/tts/de/v1_thorsten_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzäöüß .,!?…:;–'
+        example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+        jit: 'https://models.silero.ai/models/tts/de/v1_thorsten_16000.jit'
+        sample_rate: 16000
+  es:
+    v3_es:
+      latest:
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        package: 'https://models.silero.ai/models/tts/es/v3_es.pt'
+        sample_rate: [8000, 24000, 48000]
+    tux_v2:
+      latest:
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        package: 'https://models.silero.ai/models/tts/es/v2_tux.pt'
+        sample_rate: [8000, 16000]
+    tux_8khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzáéíñóú .,!?…:;–¡¿'
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        jit: 'https://models.silero.ai/models/tts/es/v1_tux_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzáéíñóú .,!?…:;–¡¿'
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        jit: 'https://models.silero.ai/models/tts/es/v1_tux_8000.jit'
+        sample_rate: 8000
+    tux_16khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzáéíñóú .,!?…:;–¡¿'
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        jit: 'https://models.silero.ai/models/tts/es/v1_tux_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzáéíñóú .,!?…:;–¡¿'
+        example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+        jit: 'https://models.silero.ai/models/tts/es/v1_tux_16000.jit'
+        sample_rate: 16000
+  fr:
+    v3_fr:
+      latest:
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        package: 'https://models.silero.ai/models/tts/fr/v3_fr.pt'
+        sample_rate: [8000, 24000, 48000]
+    gilles_v2:
+      latest:
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        package: 'https://models.silero.ai/models/tts/fr/v2_gilles.pt'
+        sample_rate: [8000, 16000]
+    gilles_8khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzéàèùâêîôûç .,!?…:;–'
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        jit: 'https://models.silero.ai/models/tts/fr/v1_gilles_8000.jit'
+        sample_rate: 8000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzéàèùâêîôûç .,!?…:;–'
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        jit: 'https://models.silero.ai/models/tts/fr/v1_gilles_8000.jit'
+        sample_rate: 8000
+    gilles_16khz:
+      latest:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzéàèùâêîôûç .,!?…:;–'
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        jit: 'https://models.silero.ai/models/tts/fr/v1_gilles_16000.jit'
+        sample_rate: 16000
+      v1:
+        tokenset: '_~abcdefghijklmnopqrstuvwxyzéàèùâêîôûç .,!?…:;–'
+        example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+        jit: 'https://models.silero.ai/models/tts/fr/v1_gilles_16000.jit'
+        sample_rate: 16000
+  ba:
+    aigul_v2:
+      latest:
+        example: 'Салауат Юлаевтың тормошо һәм яҙмышы хаҡындағы документтарҙың һәм шиғри әҫәрҙәренең бик аҙ өлөшө генә һаҡланған.'
+        package: 'https://models.silero.ai/models/tts/ba/v2_aigul.pt'
+        sample_rate: [8000, 16000]
+        language_name: 'bashkir'
+  xal:
+    v3_xal:
+      latest:
+        example: 'Һорвн, дөрвн күн ирәд, һазань чиңгнв. Байн Цецн хаана һорвн көвүн күүндҗәнә.'
+        package: 'https://models.silero.ai/models/tts/xal/v3_xal.pt'
+        sample_rate: [8000, 24000, 48000]
+    erdni_v2:
+      latest:
+        example: 'Һорвн, дөрвн күн ирәд, һазань чиңгнв. Байн Цецн хаана һорвн көвүн күүндҗәнә.'
+        package: 'https://models.silero.ai/models/tts/xal/v2_erdni.pt'
+        sample_rate: [8000, 16000]
+        language_name: 'kalmyk'
+  tt:
+    v3_tt:
+      latest:
+        example: 'Исәнмесез, саумысез, нишләп кәҗәгезне саумыйсыз, әтәчегез күкәй салган, нишләп чыгып алмыйсыз.'
+        package: 'https://models.silero.ai/models/tts/tt/v3_tt.pt'
+        sample_rate: [8000, 24000, 48000]
+    dilyara_v2:
+      latest:
+        example: 'Ис+әнмесез, с+аумысез, нишл+әп кәҗәгезн+е с+аумыйсыз, әтәчег+ез күк+әй салг+ан, нишл+әп чыг+ып +алмыйсыз.'
+        package: 'https://models.silero.ai/models/tts/tt/v2_dilyara.pt'
+        sample_rate: [8000, 16000]
+        language_name: 'tatar'
+  uz:
+    v3_uz:
+      latest:
+        example: 'Tanishganimdan xursandman.'
+        package: 'https://models.silero.ai/models/tts/uz/v3_uz.pt'
+        sample_rate: [8000, 24000, 48000]
+    dilnavoz_v2:
+      latest:
+        example: 'Tanishganimdan xursandman.'
+        package: 'https://models.silero.ai/models/tts/uz/v2_dilnavoz.pt'
+        sample_rate: [8000, 16000]
+        language_name: 'uzbek'
+  ua:
+    v3_ua:
+      latest:
+        example: 'К+отики - пухн+асті жив+отики.'
+        package: 'https://models.silero.ai/models/tts/ua/v3_ua.pt'
+        sample_rate: [8000, 24000, 48000]
+    mykyta_v2:
+      latest:
+        example: 'К+отики - пухн+асті жив+отики.'
+        package: 'https://models.silero.ai/models/tts/ua/v22_mykyta_48k.pt'
+        sample_rate: [8000, 24000, 48000]
+        language_name: 'ukrainian'
+  indic:
+    v3_indic:
+      latest:
+        example: 'prasidda kabīra adhyētā, puruṣōttama agravāla kā yaha śōdha ālēkha, usa rāmānaṁda kī khōja karatā hai'
+        package: 'https://models.silero.ai/models/tts/indic/v3_indic.pt'
+        sample_rate: [8000, 24000, 48000]
+  multi:
+    multi_v2:
+      latest:
+        package: 'https://models.silero.ai/models/tts/multi/v2_multi.pt'
+        sample_rate: [8000, 16000]
+        speakers:
+          aidar:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          baya:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          kseniya:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          irina:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          ruslan:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          natasha:
+            lang: 'ru'
+            example: 'Съ+ешьте ещ+ё +этих м+ягких франц+узских б+улочек, д+а в+ыпейте ч+аю.'
+          thorsten:
+            lang: 'de'
+            example: 'Fischers Fritze fischt frische Fische, Frische Fische fischt Fischers Fritze.'
+          tux:
+            lang: 'es'
+            example: 'Hoy ya es ayer y ayer ya es hoy, ya llegó el día, y hoy es hoy.'
+          gilles:
+            lang: 'fr'
+            example: 'Je suis ce que je suis, et si je suis ce que je suis, qu’est ce que je suis.'
+          lj:
+            lang: 'en'
+            example: 'Can you can a canned can into an un-canned can like a canner can can a canned can into an un-canned can?'
+          dilyara:
+            lang: 'tt'
+            example: 'Пес+и пес+и песик+әй, борыннар+ы бәләк+әй.'
+te_models:
+  latest:
+    package: "https://models.silero.ai/te_models/v2_4lang_q.pt"
+    languages: ['en', 'de', 'ru', 'es']
+    punct: '.,-!?—'
+  v2:
+    package: "https://models.silero.ai/te_models/v2_4lang_q.pt"
+    languages: ['en', 'de', 'ru', 'es']
+    punct: '.,-!?—'

notebooks/audiobook_gen_silero.ipynb

@@ -164,7 +164,7 @@
     "\n",
     "    ebook_title = book.get_metadata('DC', 'title')[0][0]\n",
     "    ebook_title = ebook_title.lower().replace(' ', '_')\n",
-    "    \n",
+    "\n",
     "    corpus = []\n",
     "    for item in tqdm(list(book.get_items())):\n",
     "        if item.get_type() == ebooklib.ITEM_DOCUMENT:\n",
@@ -229,7 +229,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ebook[0][:9]"
+    "ebook[0][0]"
    ]
   },
   {
@@ -249,7 +249,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "os.mkdir(f'outputs/{title}')\n",
+    "#os.mkdir(f'outputs/{title}')\n",
     "\n",
     "for chapter in tqdm(ebook[0:3]):\n",
     "    chapter_index = f'chapter{ebook.index(chapter):03}'\n",
@@ -263,11 +263,11 @@
     "        else:\n",
     "            print(f'Tensor for sentence is not valid: \\n {sentence}')\n",
     "\n",
-    "    sample_path = f'outputs/{title}/{chapter_index}.mp3'\n",
+    "    sample_path = f'outputs/{title}/{chapter_index}.wav'\n",
     "\n",
     "    if len(audio_list) > 0:\n",
     "        audio_file = torch.cat(audio_list).reshape(1, -1)\n",
-    "        torchaudio.save(sample_path, audio_file, sample_rate, format=\"mp3\")\n",
+    "#         torchaudio.save(sample_path, audio_file, sample_rate)\n",
     "    else:\n",
     "        print(f'Chapter {chapter_index} is empty.')"
    ]
@@ -313,7 +313,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },

outputs/outputs.txt

@@ -0,0 +1 @@
+blank file - this directory will contain output files

requirements.txt

@@ -3,6 +3,5 @@ streamlit
 ebooklib
 PyPDF2
 bs4
-inflect
 nltk
 stqdm

src/file_readers.py

@@ -49,9 +49,6 @@ def preprocess_text(file):
 
         sentence_list = []
         for sentence in sentences:
-            if any(chr.isdigit() for chr in sentence):
-                sentence = extract_replace(sentence)
-            sentence = replace_symbols(sentence)
             if not re.search('[a-zA-Z]', sentence):
                 sentence = ''
             wrapped_sentences = wrapper.wrap(sentence)
@@ -62,86 +59,6 @@ def preprocess_text(file):
     return text_list
 
 
-def extract_replace(entry_string):
-    import inflect
-    
-    result = (entry_string + '.')[:-1]
-    p = inflect.engine()
-    i = 0 
-
-    #initialize array with three random numbers to enter the loop, then find if there are numbers or not.
-    array = [3 , 2 , 3]
-
-    #take every number from the entry string, locate and store the number in digits in a sentence (using find_num_index), apply number_to_words
-    #to that number specifically then replace it back in the sentence.
-    while(len(array) > 2):
-        #update array with first and last indexes of every number in digits in a sentence
-        array = find_num_index(result)
-        number = result[array[i] : array[i+1] + 1]
-        k = p.number_to_words(number)
-        position = array[i]
-        number_of_characters = array[i+1] - array[i] + 1
-
-        #update sentence with the new word to numbers until there are no numbers in digits left
-        result = result[:position] + k + result[position + number_of_characters:]
-
-    return result
-
-
-def find_num_index(entry_string): 
-    result0 = []
-
-    #fill result0 array with all the indexes of digit characters in a sentence
-    for i in range(len(entry_string)):
-        if (entry_string[i].isdigit() == True):
-            result0.append(i)
-
-    result1 = []
-
-    try:
-        result1.append(result0[0])
-    except IndexError:
-        result0 = 'null'
-    if(result0 != 'null'):
-
-    # append only indexes of first and last characters of numbers to result1 array 
-        for k in range(len(result0) - 1):
-            if ((result0[k+1] - result0[k]) > 2):
-                result1.append(result0[k])
-                result1.append(result0[k+1])
-        try:
-            result1.append(result0[len(result0) - 1])
-        except IndexError:
-            result1 = 'null'
-
-    # return array of even length that contains first and last index of every number in a sentence
-    return result1
-
-
-def replace_symbols(text):
-    import re
-    
-    symbol_map = {
-        '+': ' plus ',
-        '-': ' minus ',
-        '—': ' dash ',
-        '=': ' equals ',
-        '≈': ' approximately equal to ',
-        '*': ' times ',
-        '%': ' percent ',
-        '/': ' divided by ',
-        '#': ' number ',
-        '@': ' at ',
-        '&': ' ampersand ',
-        '°': ' degrees '
-    }
-    
-    symbol_regex = re.compile('|'.join(re.escape(key) for key in symbol_map.keys()))
-    text = symbol_regex.sub(lambda x: symbol_map[x.group()], text)
-                              
-    return text
-
-
 def read_pdf(file):
     """
     Invokes PyPDF2 PdfReader to extract main body text from PDF file_like input,
@@ -189,16 +106,10 @@ def read_epub(file):
         title of document, used to name output files
 
     """
-    from pathlib import Path
-    
     import ebooklib
     from ebooklib import epub
-    from tempfile import NamedTemporaryFile
 
-    with NamedTemporaryFile(dir='.', suffix='.csv') as f:
-        f.write(file.getbuffer())
-        file = Path(f.name)
-        book = epub.read_epub(file)
+    book = epub.read_epub(file)
     file_title = book.get_metadata('DC', 'title')[0][0]
     file_title = file_title.lower().replace(' ', '_')
     corpus = []

src/output.py

@@ -35,7 +35,7 @@ def write_audio(audio_list, sample_path):
 
     if len(audio_list) > 0:
         audio_file = torch.cat(audio_list).reshape(1, -1)
-        torchaudio.save(sample_path, audio_file, cf.SAMPLE_RATE, format="mp3")
+        torchaudio.save(sample_path, audio_file, cf.SAMPLE_RATE)
         logging.info(f'Audio generated at: {sample_path}')
     else:
         logging.info(f'Audio at: {sample_path} is empty.')
@@ -67,7 +67,7 @@ def assemble_zip(title):
 
     with zipfile.ZipFile(zip_name, mode="w") as archive:
         for file_path in stqdm(config.output_path.iterdir()):
-            if file_path.suffix == '.mp3':
+            if file_path.suffix == '.wav':
                 archive.write(file_path, arcname=file_path.name)
                 file_path.unlink()
 

src/parser.py

@@ -0,0 +1,44 @@
+def read_txt(txt_path):
+    # function to read in txt files here.
+    print("Nothing here yet.")
+
+def read_epub(ebook_path):
+    import ebooklib
+    from ebooklib import epub
+    from bs4 import BeautifulSoup
+    from nltk import tokenize, download
+    from textwrap import TextWrapper
+    from stqdm import stqdm
+
+    max_char_len = 150
+
+    download('punkt', quiet=True)
+    wrapper = TextWrapper(max_char_len, fix_sentence_endings=True)
+
+    book = epub.read_epub(ebook_path)
+
+    ebook_title = book.get_metadata('DC', 'title')[0][0]
+    ebook_title = ebook_title.lower().replace(' ', '_')
+
+    corpus = []
+    for item in stqdm(list(book.get_items()), desc="Chapters in ebook:"):
+        if item.get_type() == ebooklib.ITEM_DOCUMENT:
+            input_text = BeautifulSoup(item.get_content(), "html.parser").text
+            text_list = []
+            for paragraph in input_text.split('\n'):
+                paragraph = paragraph.replace('—', '-')
+                sentences = tokenize.sent_tokenize(paragraph)
+
+                # Truncate sentences to maximum character limit
+                sentence_list = []
+                for sentence in sentences:
+                    wrapped_sentences = wrapper.wrap(sentence)
+                    sentence_list.append(wrapped_sentences)
+                # Flatten list of list of sentences
+                trunc_sentences = [phrase for sublist in sentence_list for phrase in sublist]
+
+                text_list.append(trunc_sentences)
+            text_list = [text for sentences in text_list for text in sentences]
+            corpus.append(text_list)
+
+    return corpus, ebook_title

src/predict.py

@@ -106,5 +106,5 @@ def predict(text_section, section_index, title, model, speaker):
         else:
             logging.info(f'Tensor for sentence is not valid: \n {sentence}')
 
-    sample_path = config.output_path / f'{title}_{section_index}.mp3'
+    sample_path = config.output_path / f'{title}_{section_index}.wav'
     return audio_list, sample_path