SOTA Open Source TTS
Optimize amp training
Lengyue3d002fd18b2b
diff --git a/fish_speech/models/text2semantic/llama.py b/fish_speech/models/text2semantic/llama.pyindex 41efdce..fea0933 100644--- a/fish_speech/models/text2semantic/llama.py+++ b/fish_speech/models/text2semantic/llama.py@@ -248,7 +248,13 @@ class Transformer(nn.Module):codebook_logits = self.fast_output(fast_out)# Re-pad the codebook_logits- buffer = torch.zeros(x_bs, x_len, codebook_logits.size(-1), device=x.device)+ buffer = torch.zeros(+ x_bs,+ x_len,+ codebook_logits.size(-1),+ device=codebook_logits.device,+ dtype=codebook_logits.dtype,+ )buffer[~codebook_mask] = codebook_logitscodebook_logits = bufferdiff --git a/tools/llama/generate.py b/tools/llama/generate.pyindex bb6b8ee..b2958de 100644--- a/tools/llama/generate.py+++ b/tools/llama/generate.py@@ -509,7 +509,7 @@ def main(use_prompt = prompt_text is not None and prompt_tokens is not Noneencoded = []- texts = split_text(text, 20) if iterative_prompt else [text]+ texts = split_text(text, 30) if iterative_prompt else [text]for idx, text in enumerate(texts):encoded.append(encode_tokens(@@ -561,7 +561,26 @@ def main(while seg_idx < len(encoded):seg = encoded[seg_idx]global_encoded.append(seg)- cat_encoded = torch.cat(global_encoded, dim=1)++ lengths = reversed([seg.size(1) for seg in global_encoded])+ # Pick last 2000 tokens+ count = 0+ for i, length in enumerate(lengths):+ count += length+ if count >= 2000:+ break++ if i != 0 and i % 2 == 0:+ i -= 1++ if i < len(global_encoded) - 2:+ partial_encoded = global_encoded[-i:]+ print(f"Loaded partial encoded")+ else:+ partial_encoded = global_encoded+ print(f"Using full encoded")++ cat_encoded = torch.cat(partial_encoded, dim=1)prompt_length = cat_encoded.size(1)t0 = time.perf_counter()