def GetAsrResult(self):
#本地文件方式请求
try:
cred = credential.Credential(self.SecretID, self.SecretKey)
httpProfile = HttpProfile()
httpProfile.endpoint = self.Url
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
#读取文件以及 base64
with open(self.AudioFile, "rb") as f:
audio = f.read()
if self.Format == 'pcm':
audio = utils.PcmToWav(audio)
if sys.version_info[0] == 2:
content = base64.b64encode(audio)
else:
content = base64.b64encode(audio).decode('utf-8')
#发送请求
req = models.SentenceRecognitionRequest()
params = {"ProjectId": 0, "SubServiceType": 2, "SourceType": 1, "UsrAudioKey": "session-123"}
req._deserialize(params)
req.DataLen = len(content)
req.Data = content
req.EngSerViceType = "16k_zh"
req.VoiceFormat = "wav"
req.FilterPunc = 2 #是否过滤标点符号（目前支持中文普通话引擎）。 0：不过滤，1：过滤句末标点，2：过滤所有标点。默认为0。
req.ConvertNumMode = 0 #是否进行阿拉伯数字智能转换。0：不转换，直接输出中文数字，1：根据场景智能转换为阿拉伯数字。默认值为1
resp = client.SentenceRecognition(req)
logger.info("file:%s, result:%s, sid:%s", self.AudioFile, resp.Result, resp.RequestId)
return resp.Result
except TencentCloudSDKException as err:
logger.error("code:%s, message:%s, sid:%s", err.code, err.message, err.requestId)
return ""
def get_results(self, task_id, ind):
try:
print("Getting result for %s" % self.temp_data["file_path"][ind])
cred = credential.Credential(self._secret_id, self._secret_key)
httpProfile = HttpProfile()
httpProfile.endpoint = self._endpoint
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.DescribeTaskStatusRequest()
params = {"TaskId": task_id}
req.from_json_string(json.dumps(params))
seconds = 0
while True:
time.sleep(1)
seconds += 1
resp = client.DescribeTaskStatus(req)
result_json = json.loads(resp.to_json_string())
result_status = result_json["Data"]["StatusStr"]
if result_status in ["success", "failed"]:
break
print(seconds)
result_data = result_json["Data"]["ResultDetail"]
result_data_str = json.dumps(result_data, ensure_ascii=False)
self.temp_data["result_data"].append(result_data_str)
except TencentCloudSDKException as err:
print(err)
def stt(self, audio_file):
try:
fwave = open(audio_file, mode='rb')
data = fwave.read()
dataLen = len(data)
base64Wav = base64.b64encode(data)
fwave.close()
params = '{"ProjectId":0,"SubServiceType":2,"EngSerViceType":"16k","SourceType":1,"VoiceFormat":"mp3","UsrAudioKey":"session-123", ' + '"Data":"' + str(
base64Wav, 'utf-8') + '", "DataLen":' + str(dataLen) + '}'
cred = credential.Credential(self.app_id, self.app_key)
httpProfile = HttpProfile()
httpProfile.endpoint = self.api_url
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "ap-chengdu", clientProfile)
req = models.SentenceRecognitionRequest()
req.from_json_string(params)
res = client.SentenceRecognition(req)
resp = json.loads(res.to_json_string())
return resp['Result']
except TencentCloudSDKException as err:
ASRServerError(err, "404")
def get(Taskid):
try:
cred = credential.Credential(config['tenxun_api']['secretId'],
config['tenxun_api']['secretKey'])
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.DescribeTaskStatusRequest()
params = {"TaskId": Taskid}
req.from_json_string(json.dumps(params))
resp = json.loads(client.DescribeTaskStatus(req).to_json_string())
if (resp['Data']['StatusStr'] == "success"):
if (resp['Data']['Result']):
#print(resp['Data']['Result'])
resp_e = resp['Data']['Result'].split(' ')[-1]
return resp_e
else:
return "什么都没有听到呢"
else:
return get(Taskid)
except TencentCloudSDKException as err:
get(Taskid)
def send(mid, now_time):
try:
file = open("voice/rep/{}-{}-end.wav".format(mid, now_time),
"rb").read()
data = base64.b64encode(file).decode('utf-8')
cred = credential.Credential(config['tenxun_api']['secretId'],
config['tenxun_api']['secretKey'])
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.CreateRecTaskRequest()
params = {
"EngineModelType": "16k_zh",
"ChannelNum": 1,
"ResTextFormat": 0,
"SourceType": 1,
"Data": data
}
req.from_json_string(json.dumps(params))
resp = json.loads(client.CreateRecTask(req).to_json_string())
return resp['Data']['TaskId']
except TencentCloudSDKException as err:
print(err)
send(mid, now_time)
def __init__(self):
self.cred = credential.Credential(ID, KEY)
self.httpProfile = HttpProfile()
self.httpProfile.endpoint = "asr.tencentcloudapi.com"
self.clientProfile = ClientProfile()
self.clientProfile.httpProfile = self.httpProfile
self.clientProfile.signMethod = "TC3-HMAC-SHA256" # 签名方法v3
self.client = asr_client.AsrClient(self.cred, "", self.clientProfile)
def recognize_wav(wav_path, show_detail=True):
# 通过本地语音上传方式调用
try:
# 重要：<Your SecretId>、<Your SecretKey>需要替换成用户自己的账号信息
# 请参考接口说明中的使用步骤1进行获取。
cred = credential.Credential(tencent_secret_id, tencent_secret_key)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
# 读取文件以及base64
with reader.SaveOpen(to_mp3(wav_path), 'rb') as fwave:
data = str(fwave.read())
dataLen = len(data)
base64Wav = base64.b64encode(data)
# 发送请求
req = models.SentenceRecognitionRequest()
params = {
"ProjectId": 0,
"SubServiceType": 2,
"EngSerViceType": "16k_en",
"SourceType": 1,
"Url": "",
"VoiceFormat": "mp3",
"UsrAudioKey": "session-123",
"Data": base64Wav,
"DataLen": dataLen
}
req._deserialize(params)
resp = client.SentenceRecognition(req)
if show_detail:
print(resp.to_json_string())
# windows系统使用下面一行替换上面一行
# print(resp.to_json_string().decode('UTF-8').encode('GBK') )
words = []
for w in resp.Result.split():
if not re.match('[a-zA-W]', w[-1:]):
w = w[0:-1]
words.append(w)
return ' '.join(words).lower()
except TencentCloudSDKException as err:
print(err)
return '[ERROR]'
def __init__(self, keys: Dict[str, str]):
"""
Arguments:
keys {Dict[str, str]} -- authorization keys
requires "secret_id", "secret_key"
"""
cred = credential.Credential(keys['secret_id'], keys['secret_key'])
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
self.client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
self.lang = keys.get('lang', 'zh')
def __init__(self):
##################################### 以下为自定义设置区 ##########################################
self.my_id = '' # 自己的腾讯云secretId
self.my_key = '' # 自己的腾讯云secretKey
##################################### 以上为自定义设置区 ##########################################
cred = credential.Credential(self.my_id, self.my_key)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
self.client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
def sentence_asr(self):
# 本地文件方式请求
try:
cred = credential.Credential(self.SecretId, self.SecretKey)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
# 读取文件以及 base64
# 此处可以下载测试音频 https://asr-audio-1300466766.cos.ap-nanjing.myqcloud.com/test16k.wav
with open(self.data, "rb") as f:
content = base64.b64encode(f.read()).decode('utf-8')
# 发送请求
req = models.SentenceRecognitionRequest()
if self.type:
params = {
"ProjectId": 0,
"SubServiceType": 2,
"SourceType": 1,
"UsrAudioKey": "session-123"
}
else:
params = {
"ProjectId": 0,
"SubServiceType": 2,
"SourceType": 0,
"UsrAudioKey": "session-123"
}
req.url = self.data
req._deserialize(params)
req.EngSerViceType = "16k_zh"
req.VoiceFormat = "wav"
if self.type:
req.DataLen = len(content)
req.Data = content
resp = client.SentenceRecognition(req)
res = resp.to_json_string()
res = json.loads(res)['Result']
res = res.replace('。', '').replace('，', '').replace(',', '')
return res
except TencentCloudSDKException as err:
print('tencent返回错误:%s' % err)
async def tx_asr_result(task_id: str) -> str:
"""腾讯云 ASR"""
config = manager.config
id = config["asr"]["tx_id"]
key = config["asr"]["tx_key"]
while True:
try:
cred = credential.Credential(id, key)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.DescribeTaskStatusRequest()
req.from_json_string(dumps({"TaskId": task_id}))
resp = client.DescribeTaskStatus(req)
data = resp._serialize(allow_none=True)["Data"]
status = data["StatusStr"]
if status == "success":
# detail = data["ResultDetail"]
result = data["Result"]
# duration = data["AudioDuration"]
logger.info(f"tencent cloud sdk asr is done: {task_id} {result}")
return result.strip()
elif status == "failed":
logger.error(f'task status failed: {data["ErrorMsg"]}')
else:
logger.warning(f"task status is {status}")
continue
except TencentCloudSDKException as err:
print(err)
break
async def tx_asr(audio_data: str) -> str:
"""腾讯云 ASR"""
config = manager.config
id = config["asr"]["tx_id"]
key = config["asr"]["tx_key"]
try:
cred = credential.Credential(id, key)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.CreateRecTaskRequest()
req.from_json_string(
dumps(
{
"EngineModelType": "16k_zh",
"ChannelNum": 1,
"SpeakerDiarization": 0,
"SpeakerNumber": 1,
"ResTextFormat": 2,
"SourceType": 1,
"Data": audio_data,
}
)
)
resp = client.CreateRecTask(req)
data = resp._serialize(allow_none=True)
logger.info(f"tencent cloud sdk asr is done: {data['Data']}")
return data["Data"]["TaskId"]
except TencentCloudSDKException:
logger.exception("tencent cloud sdk error")
def upload_audio(self, file_path):
print("Uploading %s" % file_path)
try:
audio_path = file_path
with open(audio_path, "rb") as f:
audio_data = f.read()
audio_data_b64 = base64.b64encode(audio_data).decode("utf-8")
cred = credential.Credential(self._secret_id, self._secret_key)
httpProfile = HttpProfile()
httpProfile.endpoint = self._endpoint
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
client = asr_client.AsrClient(cred, "", clientProfile)
req = models.CreateRecTaskRequest()
params = {
"EngineModelType": "16k_zh_video",
"ChannelNum": 1,
"SpeakerDiarization": 1,
"SpeakerNumber": 1,
"ResTextFormat": 2,
"SourceType": 1,
"Data": audio_data_b64,
}
req.from_json_string(json.dumps(params))
resp = client.CreateRecTask(req)
result_json = json.loads(resp.to_json_string())
task_id = result_json["Data"]["TaskId"]
self.temp_data["file_path"].append(audio_path)
self.temp_data["task_id"].append(task_id)
print("Upload %s success" % file_path)
except TencentCloudSDKException as err:
print(err)
import io
import sys
if sys.version_info[0] == 3:
sys.stdout = io.TextIOWrapper(sys.stdout.buffer,encoding='utf-8')
#本地文件方式请求
try:
#重要，此处<Your SecretId><Your SecretKey>需要替换成客户自己的账号信息，获取方法：
#https://cloud.tencent.com/product/asr/getting-started
cred = credential.Credential("AKIDX1GGzR5U9avzoPSmfKl79uNaaQ17nK2T", "IJ9l1D9oBH7oHWhbEaeS6VK4XjdKkz0t")
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-shanghai", clientProfile)
#读取文件以及 base64
#此处可以下载测试音频 https://asr-audio-1300466766.cos.ap-nanjing.myqcloud.com/test16k.wav
with open('./12.mp3', "rb") as f:
if sys.version_info[0] == 2:
content = base64.b64encode(f.read())
else:
content = base64.b64encode(f.read()).decode('utf-8')
#发送请求
req = models.SentenceRecognitionRequest()
params = {"ProjectId":0,"SubServiceType":2,"SourceType":1,"UsrAudioKey":"session-123", "RequestId": '1c8400f1-913f-4938-a45f-0421b08782ad' }
req._deserialize(params)
req.DataLen = len(content)
req.Data = content
req.EngSerViceType = "16k_zh"
req.VoiceFormat = "wav"
def video_to_text(video_url, video_id):
"""
input:
video_url
video_id
output:
results
format: [{time_start:xxx,time_end:xxx,text:xxx,vid:xxx},...]
"""
from tencentcloud.common import credential
from tencentcloud.common.profile.client_profile import ClientProfile
from tencentcloud.common.profile.http_profile import HttpProfile
from tencentcloud.common.exception.tencent_cloud_sdk_exception import TencentCloudSDKException
from tencentcloud.asr.v20190614 import asr_client, models
import base64
import io
import time
import sys
if sys.version_info[0] == 3:
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
# post the audio to tencent cloud platform
try:
# <Your SecretId><Your SecretKey> of tencent cloud
cred = credential.Credential(
"AKID1w0sUvY1A4nQyJC5O1PXP43qMimcZNCY", "LWSHzeGgrmQNBk4MOiqiRu40at7TYNJg")
# set http request
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-beijing", clientProfile)
# set params and create recognition task
req = models.CreateRecTaskRequest()
params = {"ChannelNum": 1, "ResTextFormat": 0, "SourceType": 0}
req._deserialize(params)
req.EngineModelType = "16k_en"
req.Url = video_url
resp = client.CreateRecTask(req)
# set result query
req = models.DescribeTaskStatusRequest()
params = '{"TaskId":%s}' % resp.Data.TaskId
req.from_json_string(params)
# polling
resp = client.DescribeTaskStatus(req)
while resp.Data.Status != 2:
print(111)
if 'faild' in resp.Data.StatusStr:
break
print(222)
print("video ID: " + str(video_id) + " is " +
resp.Data.StatusStr, flush=True)
time.sleep(1)
resp = client.DescribeTaskStatus(req)
# process the result
print("video ID: " + str(video_id) + " is " +
resp.Data.StatusStr, flush=True)
results = []
lines = resp.Data.Result.split('\n')
for line in lines:
if len(line) < 1:
continue
time, text = line.split(']')[0][1:], line.split(']')[1][2:-1]
result_dict = {}
result_dict["time_start"], result_dict[
"time_end"] = time.split(',')[0], time.split(',')[1]
result_dict["text"] = text
result_dict["vid"] = video_id
results.append(result_dict)
return results
except TencentCloudSDKException as err:
print(err)
from tencentcloud.asr.v20190614 import asr_client, models
import base64
import io
import sys
SECRET_ID = "你的Secret ID"
SECRET_KEY = "你的Secret Key"
try:
cred = credential.Credential(SECRET_ID, SECRET_KEY)
httpProfile = HttpProfile()
httpProfile.endpoint = "asr.tencentcloudapi.com"
clientProfile = ClientProfile()
clientProfile.httpProfile = httpProfile
clientProfile.signMethod = "TC3-HMAC-SHA256"
client = asr_client.AsrClient(cred, "ap-beijing", clientProfile)
# 读取文件以及 base64
with open('./geektime-00.wav', "rb") as f:
if sys.version_info[0] == 2:
content = base64.b64encode(f.read())
else:
content = base64.b64encode(f.read()).decode('utf-8')
f.close()
# 发送请求
req = models.SentenceRecognitionRequest()
params = {
"ProjectId": 0,
"SubServiceType": 2,
"SourceType": 1,
"UsrAudioKey": "sessionid-geektime"
}
