我正在尝试从我的 Twilio 帐户/号码向 PSTN 手机号码拨打电话。我正在使用这个 TwiML：
有了这个，我就可以接收和播放来自 PSTN 手机的传入音频
我无法将传出音频发送到 PSTN 手机。
这是主要的 websocket 函数，它处理传入的音频并尝试 通过函数“sendremotemediatotwilio”发送传出数据包作为响应：
async def websocket_rpc_endpoint(ws: WebSocket):
# A lot of messages will be sent rapidly. We'll stop showing after the first one.
global output_queue
inbuffer = bytearray(b'')
inbound_chunks_started = False
latest_inbound_timestamp = 0
BUFFER_SIZE = frames_per_buffer
await ws.accept()
while True:
data = await ws.receive_json()
# Using the event type you can determine what type of message you are receiving
if data['event'] == "connected":
print("Connected Message received: {}".format(data))
if data['event'] == "start":
print("Start Message received: {}".format(data))
if data['event'] == "media":
# print("Media message: {}".format(data))
media = data['media']
chunk = base64.b64decode(media['payload'])
# print("Payload is: {}".format(payload))
if media['track'] == 'inbound':
# fills in silence if there have been dropped packets
if inbound_chunks_started:
if latest_inbound_timestamp + 20 < int(media['timestamp']):
bytes_to_fill = 8 * (int(media['timestamp']) - (latest_inbound_timestamp + 20))
# NOTE: 0xff is silence for mulaw audio
# and there are 8 bytes per ms of data for our format (8 bit, 8000 Hz)
inbuffer.extend(b'\xff' * bytes_to_fill)
else:
# make it known that inbound chunks have started arriving
inbound_chunks_started = True
latest_inbound_timestamp = int(media['timestamp'])
latest_inbound_timestamp = int(media['timestamp'])
# extend the inbound audio buffer with data
inbuffer.extend(chunk)
while len(inbuffer) >= BUFFER_SIZE:
asinbound = AudioSegment(inbuffer[:BUFFER_SIZE], sample_width=1, frame_rate=8000, channels=1)
# # print("That's {} bytes".format(len(chunk)))
# print("That's {} bytes".format(len(chunk)))
# print("Additional media messages from WebSocket are being suppressed....")
output_queue.put_nowait(base64.b64encode(asinbound.raw_data))
inbuffer = inbuffer[BUFFER_SIZE:]
await sendremotemediatotwilio(ws,media,data['streamSid'])
# print("Added data in queue")
if data['event'] == "closed":
print("Closed Message received: {}".format(data))
break
async def sendremotemediatotwilio(ws,media,streamSid):
global input_queue
# if input_queue.empty():
# await asyncio.sleep(1)
if not input_queue.empty():
base64_data = input_queue.get_nowait()
media_data = {
"event": "media",
"streamSid": streamSid,
"media": {
"payload": base64.b64encode(base64_data).decode('utf-8')
}
}
# media = json.dumps(media_data)
# print(f"media: {media}")
print("sending json")
await ws.send_json(media_data)
else:
# print("Queue is empty")
pass
我已经解决了这个问题。现在我可以成功地向 PSTN 手机发送数据包和从 PSTN 手机接收数据包。问题在于从 pcmu 转码到线性 PCM。一旦解决，相同的代码就可以正常工作了
@app.websocket("/twiliomedia")
async def websocket_rpc_endpoint(ws: WebSocket):
# A lot of messages will be sent rapidly. We'll stop showing after the first one.
global output_queue,pushdatainqueue
inbuffer = bytearray(b'')
inbound_chunks_started = False
latest_inbound_timestamp = 0
BUFFER_SIZE = frames_per_buffer
await ws.accept()
while True:
data = await ws.receive_json()
# print("Json data",data)
# if message is None:
# print("No message received...")
# continue
#
# # Messages are a JSON encoded string
# data = json.loads(message)
# Using the event type you can determine what type of message you are receiving
if data['event'] == "connected":
print("Connected Message received: {}".format(data))
if data['event'] == "start":
print("Start Message received: {}".format(data))
if data['event'] == "media":
# print("Media message: {}".format(data))
media = data['media']
chunk = base64.b64decode(media['payload'])
# print("Payload is: {}".format(payload))
if media['track'] == 'inbound':
# fills in silence if there have been dropped packets
if inbound_chunks_started:
if latest_inbound_timestamp + 20 < int(media['timestamp']):
bytes_to_fill = 8 * (int(media['timestamp']) - (latest_inbound_timestamp + 20))
# NOTE: 0xff is silence for mulaw audio
# and there are 8 bytes per ms of data for our format (8 bit, 8000 Hz)
inbuffer.extend(b'\xff' * bytes_to_fill)
else:
# make it known that inbound chunks have started arriving
inbound_chunks_started = True
pushdatainqueue = True
latest_inbound_timestamp = int(media['timestamp'])
latest_inbound_timestamp = int(media['timestamp'])
# extend the inbound audio buffer with data
inbuffer.extend(chunk)
while len(inbuffer) >= BUFFER_SIZE:
asinbound = AudioSegment(inbuffer[:BUFFER_SIZE], sample_width=1, frame_rate=8000, channels=1)
# # print("That's {} bytes".format(len(chunk)))
# print("That's {} bytes".format(len(chunk)))
# print("Additional media messages from WebSocket are being suppressed....")
output_queue.put_nowait(base64.b64encode(asinbound.raw_data))
inbuffer = inbuffer[BUFFER_SIZE:]
await sendremotemediatotwilio(ws, media, data['streamSid'])
# print("Added data in queue")
if data['event'] == "closed":
print("Closed Message received: {}".format(data))
break
async def sendremotemediatotwilio(ws, media, streamSid):
global input_queue
# if input_queue.empty():
# await asyncio.sleep(1)
if not input_queue.empty():
base64_data = input_queue.get_nowait()
media_data = {
"event": "media",
"streamSid": streamSid,
"media": {
"payload": base64.b64encode(base64_data).decode("utf-8")
# "payload": base64_data.decode("utf-8")
}
}
# media = json.dumps(media_data)
# print(f"media: {media}")
print("sending json")
await ws.send_json(media_data)
else:
# print("Queue is empty")
pass