deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-04 17:39:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix lint

Browse Source
This commit is contained in:
Yuekai Zhang
2025-09-03 03:45:17 -07:00
parent 633b991290
commit ad257b06e3
4 changed files with 14 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
runtime/triton_trtllm/client_grpc.py
View File

@@ -413,7 +413,7 @@ def run_sync_streaming_inference(
for i in range(1, len(audios)): for i in range(1, len(audios)):
# Cross-fade section # Cross-fade section
cross_faded_overlap = (audios[i][:cross_fade_samples] * fade_in + cross_faded_overlap = (audios[i][:cross_fade_samples] * fade_in +
audios[i - 1][-cross_fade_samples:] * fade_out) audios[i - 1][-cross_fade_samples:] * fade_out)
# Middle section of the current chunk # Middle section of the current chunk
middle_part = audios[i][cross_fade_samples:-cross_fade_samples] middle_part = audios[i][cross_fade_samples:-cross_fade_samples]
# Concatenate # Concatenate

19
runtime/triton_trtllm/model_repo/cosyvoice2/1/model.py
View File

@@ -41,11 +41,11 @@ from transformers import AutoTokenizer
import torchaudio import torchaudio
from matcha.utils.audio import mel_spectrogram from matcha.utils.audio import mel_spectrogram
torch.set_num_threads(1) torch.set_num_threads(1)
class TritonPythonModel: class TritonPythonModel:
"""Triton Python model for Spark TTS. """Triton Python model for Spark TTS.
@@ -65,7 +65,7 @@ class TritonPythonModel:
parameters = self.model_config['parameters'] parameters = self.model_config['parameters']
model_params = {k: v["string_value"] for k, v in parameters.items()} model_params = {k: v["string_value"] for k, v in parameters.items()}
self.logger.log_info(f"model_params:{model_params}") self.logger.log_info(f"model_params:{model_params}")
self.dynamic_chunk_strategy = model_params.get("dynamic_chunk_strategy", "exponential") # "exponential" or "time_based" self.dynamic_chunk_strategy = model_params.get("dynamic_chunk_strategy", "exponential") # "exponential" or "time_based"
self.logger.log_info(f"Using dynamic chunk strategy: {self.dynamic_chunk_strategy}") self.logger.log_info(f"Using dynamic chunk strategy: {self.dynamic_chunk_strategy}")
# Initialize tokenizer # Initialize tokenizer
@@ -193,7 +193,6 @@ class TritonPythonModel:
return prompt_speech_tokens return prompt_speech_tokens
def forward_speaker_embedding(self, wav): def forward_speaker_embedding(self, wav):
"""Forward pass through the speaker embedding component. """Forward pass through the speaker embedding component.
@@ -219,7 +218,6 @@ class TritonPythonModel:
return prompt_spk_embedding return prompt_spk_embedding
def forward_token2wav( def forward_token2wav(
self, self,
prompt_speech_tokens: torch.Tensor, prompt_speech_tokens: torch.Tensor,
@@ -254,7 +252,6 @@ class TritonPythonModel:
inputs_tensor.append(token_offset_tensor) inputs_tensor.append(token_offset_tensor)
inputs_tensor.append(finalize_tensor) inputs_tensor.append(finalize_tensor)
# Create and execute inference request # Create and execute inference request
inference_request = pb_utils.InferenceRequest( inference_request = pb_utils.InferenceRequest(
model_name='token2wav', model_name='token2wav',
@@ -281,7 +278,6 @@ class TritonPythonModel:
input_ids = torch.cat([input_ids, prompt_speech_tokens], dim=1) input_ids = torch.cat([input_ids, prompt_speech_tokens], dim=1)
return input_ids return input_ids
def _extract_speech_feat(self, speech): def _extract_speech_feat(self, speech):
speech_feat = mel_spectrogram( speech_feat = mel_spectrogram(
speech, speech,
@@ -299,7 +295,6 @@ class TritonPythonModel:
speech_feat = speech_feat.unsqueeze(dim=0) speech_feat = speech_feat.unsqueeze(dim=0)
return speech_feat return speech_feat
def _llm_gen_thread(self, generated_ids_iter, semantic_token_ids_arr, llm_is_done_flag): def _llm_gen_thread(self, generated_ids_iter, semantic_token_ids_arr, llm_is_done_flag):
for generated_ids in generated_ids_iter: for generated_ids in generated_ids_iter:
generated_ids = generated_ids.tolist() generated_ids = generated_ids.tolist()
@@ -338,9 +333,8 @@ class TritonPythonModel:
prompt_speech_feat = speech_feat[:, :2 * token_len].contiguous().half() prompt_speech_feat = speech_feat[:, :2 * token_len].contiguous().half()
prompt_speech_tokens = prompt_speech_tokens[:, :token_len].contiguous() prompt_speech_tokens = prompt_speech_tokens[:, :token_len].contiguous()
flow_prompt_speech_token_len = prompt_speech_tokens.shape[-1] flow_prompt_speech_token_len = prompt_speech_tokens.shape[-1]
reference_text = pb_utils.get_input_tensor_by_name(request, "reference_text").as_numpy() reference_text = pb_utils.get_input_tensor_by_name(request, "reference_text").as_numpy()
reference_text = reference_text[0][0].decode('utf-8') reference_text = reference_text[0][0].decode('utf-8')
@@ -385,7 +379,9 @@ class TritonPythonModel:
this_tts_speech_token = semantic_token_ids_arr[:token_offset + this_token_hop_len + self.flow_pre_lookahead_len] this_tts_speech_token = semantic_token_ids_arr[:token_offset + this_token_hop_len + self.flow_pre_lookahead_len]
this_tts_speech_token = torch.tensor(this_tts_speech_token).unsqueeze(dim=0).to(torch.int32).to(self.device) this_tts_speech_token = torch.tensor(this_tts_speech_token).unsqueeze(dim=0).to(torch.int32).to(self.device)
sub_tts_speech = self.forward_token2wav(prompt_speech_tokens, prompt_speech_feat, prompt_spk_embedding, this_tts_speech_token, request_id, token_offset, False) sub_tts_speech = self.forward_token2wav(
prompt_speech_tokens, prompt_speech_feat, prompt_spk_embedding,
this_tts_speech_token, request_id, token_offset, False)
audio_tensor = pb_utils.Tensor.from_dlpack("waveform", to_dlpack(sub_tts_speech)) audio_tensor = pb_utils.Tensor.from_dlpack("waveform", to_dlpack(sub_tts_speech))
inference_response = pb_utils.InferenceResponse(output_tensors=[audio_tensor]) inference_response = pb_utils.InferenceResponse(output_tensors=[audio_tensor])
@@ -413,7 +409,6 @@ class TritonPythonModel:
else: else:
this_token_hop_len = self.token_hop_len this_token_hop_len = self.token_hop_len
this_token_hop_len = max(self.token_hop_len, this_token_hop_len) this_token_hop_len = max(self.token_hop_len, this_token_hop_len)
chunk_index += 1 chunk_index += 1
else: else:
time.sleep(0.02) time.sleep(0.02)
@@ -423,7 +418,7 @@ class TritonPythonModel:
audio_tensor = pb_utils.Tensor.from_dlpack("waveform", to_dlpack(sub_tts_speech)) audio_tensor = pb_utils.Tensor.from_dlpack("waveform", to_dlpack(sub_tts_speech))
inference_response = pb_utils.InferenceResponse(output_tensors=[audio_tensor]) inference_response = pb_utils.InferenceResponse(output_tensors=[audio_tensor])
response_sender.send(inference_response) response_sender.send(inference_response)
llm_thread.join() llm_thread.join()
response_sender.send(flags=pb_utils.TRITONSERVER_RESPONSE_COMPLETE_FINAL) response_sender.send(flags=pb_utils.TRITONSERVER_RESPONSE_COMPLETE_FINAL)
self.logger.log_info("send tritonserver_response_complete_final to end") self.logger.log_info("send tritonserver_response_complete_final to end")

11
runtime/triton_trtllm/model_repo/speaker_embedding/1/model.py
View File

@@ -57,13 +57,13 @@ class TritonPythonModel:
self.device = torch.device("cuda") self.device = torch.device("cuda")
model_dir = model_params["model_dir"] model_dir = model_params["model_dir"]
gpu="l20" gpu = "l20"
enable_trt = True enable_trt = True
if enable_trt: if enable_trt:
self.load_spk_trt(f'{model_dir}/campplus.{gpu}.fp32.trt', self.load_spk_trt(f'{model_dir}/campplus.{gpu}.fp32.trt',
f'{model_dir}/campplus.onnx', f'{model_dir}/campplus.onnx',
1, 1,
False) False)
else: else:
campplus_model = f'{model_dir}/campplus.onnx' campplus_model = f'{model_dir}/campplus.onnx'
option = onnxruntime.SessionOptions() option = onnxruntime.SessionOptions()
@@ -121,7 +121,7 @@ class TritonPythonModel:
assert spk_model.execute_async_v3(torch.cuda.current_stream().cuda_stream) is True assert spk_model.execute_async_v3(torch.cuda.current_stream().cuda_stream) is True
torch.cuda.current_stream().synchronize() torch.cuda.current_stream().synchronize()
self.spk_model.release_estimator(spk_model, stream) self.spk_model.release_estimator(spk_model, stream)
return embedding.half() return embedding.half()
def execute(self, requests): def execute(self, requests):
@@ -142,7 +142,6 @@ class TritonPythonModel:
wav_array = torch.from_numpy(wav_array).to(self.device) wav_array = torch.from_numpy(wav_array).to(self.device)
embedding = self._extract_spk_embedding(wav_array) embedding = self._extract_spk_embedding(wav_array)
prompt_spk_embedding_tensor = pb_utils.Tensor.from_dlpack( prompt_spk_embedding_tensor = pb_utils.Tensor.from_dlpack(
"prompt_spk_embedding", to_dlpack(embedding)) "prompt_spk_embedding", to_dlpack(embedding))

2
runtime/triton_trtllm/model_repo/token2wav/1/model.py
View File

@@ -49,6 +49,7 @@ logger = logging.getLogger(__name__)
ORIGINAL_VOCAB_SIZE = 151663 ORIGINAL_VOCAB_SIZE = 151663
torch.set_num_threads(1) torch.set_num_threads(1)
class CosyVoice2: class CosyVoice2:
def __init__(self, model_dir, load_jit=False, load_trt=False, fp16=False, trt_concurrent=1, device='cuda'): def __init__(self, model_dir, load_jit=False, load_trt=False, fp16=False, trt_concurrent=1, device='cuda'):
@@ -123,7 +124,6 @@ class CosyVoice2Model:
input_names = ["x", "mask", "mu", "cond"] input_names = ["x", "mask", "mu", "cond"]
return {'min_shape': min_shape, 'opt_shape': opt_shape, 'max_shape': max_shape, 'input_names': input_names} return {'min_shape': min_shape, 'opt_shape': opt_shape, 'max_shape': max_shape, 'input_names': input_names}
def token2wav(self, token, prompt_token, prompt_feat, embedding, token_offset, uuid, stream=False, finalize=False, speed=1.0): def token2wav(self, token, prompt_token, prompt_feat, embedding, token_offset, uuid, stream=False, finalize=False, speed=1.0):
with torch.cuda.amp.autocast(self.fp16): with torch.cuda.amp.autocast(self.fp16):
tts_mel, _ = self.flow.inference(token=token.to(self.device), tts_mel, _ = self.flow.inference(token=token.to(self.device),