decoding

游雁

2024-06-13 5de8bfdcd8a617ac13c13478505401bbf4e57472

 examples/industrial_data_pretraining/llm_asr/demo_speech2text.py

@@ -3,20 +3,41 @@
# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
#  MIT License  (https://opensource.org/licenses/MIT)

import json
import os
import sys

from funasr import AutoModel

model = AutoModel(
    model="/nfs/beinian.lzr/workspace/GPT-4o/Exp/exp6/4m-8gpu/exp6_speech2text_0607_linear_ddp",
)

ckpt_dir = "/nfs/beinian.lzr/workspace/GPT-4o/Exp/exp6/5m-8gpu/exp6_speech2text_linear_ddp_0609"
ckpt_id = "model.pt.ep0.90000"
jsonl = (
    "/nfs/beinian.lzr/workspace/GPT-4o/Data/Speech2Text/TestData/aishell1_test_speech2text.jsonl"
)
output_dir = f"{os.path.join(ckpt_dir, ckpt_id)}"
device = "cuda:0"

ckpt_dir = sys.argv[1]
ckpt_id = sys.argv[2]
jsonl = sys.argv[3]
output_dir = sys.argv[4]
device = sys.argv[5]

model = AutoModel(
    model=ckpt_dir,
    init_param=f"{os.path.join(ckpt_dir, ckpt_id)}",
    output_dir=output_dir,
    device=device,
    fp16=False,
    bf16=False,
    llm_dtype="bf16",
)


with open(jsonl, "r") as f:
    lines = f.readlines()

tearchforing = True
tearchforing = False
for i, line in enumerate(lines):
    data_dict = json.loads(line.strip())
    data = data_dict["messages"]