init

.gitignore

@@ -0,0 +1,26 @@
+# General
+.DS_Store
+.AppleDouble
+.LSOverride
+
+# Icon must end with two \r
+Icon
+
+# Thumbnails
+._*
+
+# Files that might appear in the root of a volume
+.DocumentRevisions-V100
+.fseventsd
+.Spotlight-V100
+.TemporaryItems
+.Trashes
+.VolumeIcon.icns
+.com.apple.timemachine.donotpresent
+
+# Directories potentially created on remote AFP share
+.AppleDB
+.AppleDesktop
+Network Trash Folder
+Temporary Items
+.apdisk

khm/G_100000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86a7237c6bd72beb885152631e69743290c03b4a33ea17b7c3ca851a29b9749d
+size 436464113

khm/config.json

@@ -0,0 +1,87 @@
+{
+    "train": {
+        "log_interval": 200,
+        "eval_interval": 1000,
+        "seed": 1234,
+        "epochs": 20000,
+        "learning_rate": 0.0002,
+        "betas": [
+            0.8,
+            0.99
+        ],
+        "eps": 1e-09,
+        "batch_size": 64,
+        "fp16_run": true,
+        "lr_decay": 0.999875,
+        "segment_size": 8192,
+        "init_lr_ratio": 1,
+        "warmup_epochs": 0,
+        "c_mel": 45,
+        "c_kl": 1.0
+    },
+    "data": {
+        "training_files": "train.ltr",
+        "validation_files": "dev.ltr",
+        "text_cleaners": [
+            "transliteration_cleaners"
+        ],
+        "max_wav_value": 32768.0,
+        "sampling_rate": 16000,
+        "filter_length": 1024,
+        "hop_length": 256,
+        "win_length": 1024,
+        "n_mel_channels": 80,
+        "mel_fmin": 0.0,
+        "mel_fmax": null,
+        "add_blank": true,
+        "n_speakers": 0,
+        "cleaned_text": true
+    },
+    "model": {
+        "inter_channels": 192,
+        "hidden_channels": 192,
+        "filter_channels": 768,
+        "n_heads": 2,
+        "n_layers": 6,
+        "kernel_size": 3,
+        "p_dropout": 0.1,
+        "resblock": "1",
+        "resblock_kernel_sizes": [
+            3,
+            7,
+            11
+        ],
+        "resblock_dilation_sizes": [
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ]
+        ],
+        "upsample_rates": [
+            8,
+            8,
+            2,
+            2
+        ],
+        "upsample_initial_channel": 512,
+        "upsample_kernel_sizes": [
+            16,
+            16,
+            4,
+            4
+        ],
+        "n_layers_q": 3,
+        "use_spectral_norm": false
+    }
+}

khm/vocab.txt

@@ -0,0 +1,74 @@
+្
+យ
+អ
+ឆ
+គ
+ង
+ើ
+ធ
+់
+ឃ
+ឌ
+ឥ
+ប
+ួ
+ឈ
+q
+៏
+ផ
+ិ
+ 
+ជ
+ឧ
+ល
+ឱ
+ទ
+ូ
+រ
+៎
+ឭ
+ា
+ៀ
+ឿ
+េ
+ំ
+៍
+ត
+_
+៉
+ក
+ែ
+ៅ
+ៃ
+ដ
+វ
+ៈ
+ឬ
+ឫ
+ះ
+ោ
+ី
+ុ
+ឯ
+ព
+ឡ
+ច
+ឋ
+ណ
+1
+ហ
+ន
+ឹ
+ស
+ខ
+័
+ញ
+៊
+ឮ
+៌
+ថ
+ឺ
+ឪ
+-
+ម
+ភ

main.py

@@ -0,0 +1,44 @@
+# -*- coding: utf-8 -*-
+import re
+from ttsmms import TTS
+import gradio as gr
+
+tts = TTS("khm")
+
+def sanitize(text):
+    return re.sub(r"\u200b", "", text)
+
+def generate_voice(text):
+    audio = tts.synthesis(text)
+    return (audio['sampling_rate'], audio['x'])
+
+with gr.Blocks(title="Khmer Text to Speech with MMS") as blocks:
+
+    gr.Markdown('# Khmer Text to Speech - MMS')
+    gr.Markdown('MMS: Scaling Speech Technology to 1000+ languages by Meta AI')
+
+    input_text = gr.Textbox(label="អក្សរ", lines=3)
+    examples = gr.Examples(examples=["នេះ​ជា​រូបថត​ជនសង្ស័យ​ដែល​បេីកឡាន។"], inputs=[input_text])
+
+    run_button = gr.Button(
+        text="Generate", 
+        type="button",
+    )
+
+    out_audio = gr.Audio(
+        label="សំឡេងដែលបានបង្កើត",
+        type="numpy", 
+    )
+
+    inputs = [input_text]
+    outputs = [out_audio]
+
+    run_button.click(
+        fn=generate_voice,
+        inputs=inputs,
+        outputs=outputs,
+        queue=True,
+    )
+
+
+blocks.queue(concurrency_count=1).launch(debug=True)

packages.txt

@@ -0,0 +1,2 @@
+libsndfile1
+espeak-ng

requirements.txt

@@ -0,0 +1,2 @@
+ttsmms==0.6
+gradio==3.32.0