[Example] ChatTTS: add example

Author: grorge123

wasmedge-chatTTS/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name = "wasmedge-chattts"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+serde_json = "1.0"
+wasmedge-wasi-nn = {path = "../../wasmedge-wasi-nn/rust", version = "0.8.0"}
+hound = "3.4"

wasmedge-chatTTS/README.md

@@ -0,0 +1,3 @@
+cargo build --target wasm32-wasi --release
+
+wasmedge --dir .:.  ./target/wasm32-wasi/release/wasmedge-chattts.wasm

wasmedge-chatTTS/config/decoder.yaml

@@ -0,0 +1,12 @@
+
+
+dim: 384
+
+decoder_config:
+  idim: ${dim}
+  odim: ${dim}
+  hidden: 512
+  n_layer: 12
+  bn_dim: 128
+
+vq_config: null

wasmedge-chatTTS/config/dvae.yaml

@@ -0,0 +1,14 @@
+
+
+dim: 512
+decoder_config:
+  idim: ${dim}
+  odim: ${dim}
+  n_layer: 12
+  bn_dim: 128
+
+vq_config:
+  dim: 1024
+  levels: [5,5,5,5]
+  G: 2
+  R: 2

wasmedge-chatTTS/config/gpt.yaml

@@ -0,0 +1,20 @@
+
+
+num_audio_tokens: 626
+num_text_tokens: 21178
+
+gpt_config:
+  hidden_size: 768
+  intermediate_size: 3072
+  num_attention_heads: 12
+  num_hidden_layers: 20
+  use_cache: False
+  max_position_embeddings: 4096
+  # attn_implementation: flash_attention_2
+
+  spk_emb_dim: 192
+  spk_KL: False
+  num_audio_tokens: 626
+  num_text_tokens: null
+  num_vq: 4
+

wasmedge-chatTTS/config/path.yaml

@@ -0,0 +1,11 @@
+
+
+vocos_config_path: config/vocos.yaml
+vocos_ckpt_path: asset/Vocos.pt
+dvae_config_path: config/dvae.yaml
+dvae_ckpt_path: asset/DVAE.pt
+gpt_config_path: config/gpt.yaml
+gpt_ckpt_path: asset/GPT.pt
+decoder_config_path: config/decoder.yaml
+decoder_ckpt_path: asset/Decoder.pt
+tokenizer_path: asset/tokenizer.pt

wasmedge-chatTTS/config/vocos.yaml

@@ -0,0 +1,24 @@
+feature_extractor:
+  class_path: vocos.feature_extractors.MelSpectrogramFeatures
+  init_args:
+    sample_rate: 24000
+    n_fft: 1024
+    hop_length: 256
+    n_mels: 100
+    padding: center
+
+backbone:
+  class_path: vocos.models.VocosBackbone
+  init_args:
+    input_channels: 100
+    dim: 512
+    intermediate_dim: 1536
+    num_layers: 8
+
+head:
+  class_path: vocos.heads.ISTFTHead
+  init_args:
+    dim: 512
+    n_fft: 1024
+    hop_length: 256
+    padding: center

wasmedge-chatTTS/src/main.rs

@@ -0,0 +1,52 @@
+use wasmedge_wasi_nn::{
+    self, ExecutionTarget, GraphBuilder, GraphEncoding, GraphExecutionContext,
+    TensorType,
+};
+use hound;
+
+fn get_data_from_context(context: &GraphExecutionContext, index: usize, limit: usize) -> Vec<u8> {
+    // Preserve for 4096 tokens with average token length 8
+    const MAX_OUTPUT_BUFFER_SIZE: usize = 4096 * 4096;
+    let mut output_buffer = vec![0u8; MAX_OUTPUT_BUFFER_SIZE];
+    let _ = context
+        .get_output(index, &mut output_buffer)
+        .expect("Failed to get output");
+
+    return output_buffer[..limit].to_vec();
+}
+
+fn main() {
+    let prompt = "It is a test sentence.";
+    let tensor_data = prompt.as_bytes().to_vec();
+    let empty_vec: Vec<Vec<u8>> = Vec::new();
+    let graph = GraphBuilder::new(GraphEncoding::ChatTTS, ExecutionTarget::CPU)
+        .build_from_bytes(empty_vec)
+        .expect("Failed to build graph");
+    let mut context = graph
+        .init_execution_context()
+        .expect("Failed to init context");
+    context
+        .set_input(0, TensorType::U8, &[1], &tensor_data)
+        .expect("Failed to set input");
+    context.compute().expect("Failed to compute");
+    let bytes_written = get_data_from_context(&context, 1, 4);
+    let bytes_written = usize::from_le_bytes(bytes_written.as_slice().try_into().unwrap());
+    println!("Byte: {}", bytes_written);
+    let output_bytes = get_data_from_context(&context, 0, bytes_written);
+    let spec = hound::WavSpec {
+        channels: 1,
+        sample_rate: 24000,
+        bits_per_sample: 32,
+        sample_format: hound::SampleFormat::Float,
+    };
+    let mut writer = hound::WavWriter::create("output1.wav", spec).unwrap();
+    let samples: Vec<f32> = output_bytes
+        .chunks_exact(4)
+        .map(|b| f32::from_le_bytes([b[0], b[1], b[2], b[3]]))
+        .collect();
+    for sample in samples {
+        writer.write_sample(sample).unwrap();
+    }
+    writer.finalize().unwrap();
+    graph.unload().expect("Failed to free resource");
+}