Files
edge-tts/src/edgeTTS/__init__.py
2021-06-08 19:16:13 +03:00

260 lines
13 KiB
Python
Executable File

#!/usr/bin/env python3
import sys
import json
import uuid
import argparse
import asyncio
import ssl
import websockets
import logging
import httpx
import time
import math
from xml.sax.saxutils import escape
# Default variables
ssl_context = ssl.create_default_context()
trustedClientToken = '6A5AA1D4EAFF4E9FB37E23D68491D6F4'
wssUrl = 'wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=' + trustedClientToken
voiceList = 'https://speech.platform.bing.com/consumer/speech/synthesize/readaloud/voices/list?trustedclienttoken=' + trustedClientToken
# Return date format in Microsoft Edge's broken way (Edge does it wrong because they
# append Z to a date with locale time zone). They probably just use Date().toString()
def formatdate():
return time.strftime('%a %b %d %Y %H:%M:%S GMT+0000 (Coordinated Universal Time)', time.gmtime())
def bool_to_lower_str(x):
return 'true' if x else 'false'
def connectId():
return str(uuid.uuid4()).replace("-", "")
def removeIncompatibleControlChars(s):
logger = logging.getLogger("edgeTTS.removeIncompatibleControlChars")
output = ""
for char in s:
char_code = ord(char)
if (char_code >= 0 and char_code <= 8) or (char_code >= 11 and char_code <= 12) \
or (char_code >= 14 and char_code <= 31):
logger.debug("Forbidden character %s" % char.encode('utf-8'))
output += ' '
else:
logger.debug("Allowed character %s" % char.encode('utf-8'))
output += char
logger.debug("Generated %s" % output.encode('utf-8'))
return output
def mktimestamp(ns):
hour = math.floor(ns / 10000 / 1000 / 3600)
minute = math.floor((ns / 10000 / 1000 / 60) % 60)
seconds = math.floor((ns / 10000 / 1000) % 60)
mili = float(str(math.modf((ns / 10000) - (1000 * seconds))[1])[:3])
return "%.02d:%.02d:%.02d.%.03d" % (hour, minute, seconds, mili)
def list_voices():
logger = logging.getLogger("edgeTTS.list_voices")
with httpx.Client(http2=True, headers={
'Authority': 'speech.platform.bing.com',
'Sec-CH-UA': "\" Not;A Brand\";v=\"99\", \"Microsoft Edge\";v=\"91\", \"Chromium\";v=\"91\"",
'Sec-CH-UA-Mobile': '?0',
'User-Agent': "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.77 Safari/537.36 Edg/91.0.864.41",
'Accept': '*/*',
'Sec-Fetch-Site': 'none',
'Sec-Fetch-Mode': 'cors',
'Sec-Fetch-Dest': 'empty',
'Accept-Encoding': 'gzip, deflate, br',
'Accept-Language': 'en-US,en;q=0.9'
}) as url:
logger.debug("Loading json from %s" % voiceList)
data = json.loads(url.get(voiceList).content)
logger.debug("JSON Loaded")
return data
class SubMaker:
def __init__(self, overlapping=5):
self.subsAndOffset = {}
self.overlapping = (overlapping * (10**7))
def formatter(self, offset1, offset2, subdata):
data = "%s --> %s\r\n" % (mktimestamp(offset1), mktimestamp(offset2))
data += "%s\r\n\r\n" % escape(subdata)
return data
def createSub(self, timestamp, text):
self.subsAndOffset.update({ timestamp: text })
def generateSubs(self):
oldTimeStamp = None
oldSubData = None
data = "WEBVTT\r\n\r\n"
first = sorted(self.subsAndOffset.keys(), key=int)[0]
try:
second = sorted(self.subsAndOffset.keys(), key=int)[1]
data += self.formatter(first, second + self.overlapping, self.subsAndOffset[first]) ## overlapping Subtitles
except IndexError: # This means TTS said one word only.
data += self.formatter(0, first + ((10**7) * 10), self.subsAndOffset[first])
for sub in sorted(self.subsAndOffset.keys(), key=int)[1:]:
if (oldTimeStamp and oldSubData) is not None:
data += self.formatter(oldTimeStamp, sub + self.overlapping, oldSubData) ## overlapping Subtitles
oldTimeStamp = sub
oldSubData = self.subsAndOffset[sub]
if (oldTimeStamp and oldSubData) is not None:
data += self.formatter(oldTimeStamp, oldTimeStamp + ((10**7) * 10), oldSubData)
return data
class Communicate:
def __init__(self):
self.date = formatdate()
def mkssmlmsg(self, text="", voice="", pitch="", rate="", volume="", customspeak=False):
message='X-RequestId:'+connectId()+'\r\nContent-Type:application/ssml+xml\r\n'
message+='X-Timestamp:'+self.date+'Z\r\nPath:ssml\r\n\r\n'
if customspeak:
message+=text
else:
message+="<speak version='1.0' xmlns='http://www.w3.org/2001/10/synthesis' xml:lang='en-US'>"
message+="<voice name='" + voice + "'>" + "<prosody pitch='" + pitch + "' rate ='" + rate + "' volume='" + volume + "'>" + text + '</prosody></voice></speak>'
return message
async def run(self, msg, sentenceBoundary=False, wordBoundary=False, codec="audio-24khz-48kbitrate-mono-mp3", voice="Microsoft Server Speech Text to Speech Voice (en-US, AriaNeural)", pitch="+0Hz", rate="+0%", volume="+0%", customspeak=False):
sentenceBoundary = bool_to_lower_str(sentenceBoundary)
wordBoundary = bool_to_lower_str(wordBoundary)
if not customspeak:
wsmax = 2 ** 16
overhead = len(self.mkssmlmsg("", voice, pitch, rate, volume, customspeak=False))
msgs = _minimize(escape(removeIncompatibleControlChars(msg)), b" ", wsmax - overhead)
async with websockets.connect(
wssUrl + "&ConnectionId=" + connectId(),
ssl=ssl_context,
compression="deflate",
extra_headers={
"Pragma": "no-cache",
"Origin": "chrome-extension://jdiccldimpdaibmpdkjnbmckianbfold",
"Accept-Encoding": "gzip, deflate, br",
"Accept-Language": "en-US,en;q=0.9",
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.77 Safari/537.36 Edg/91.0.864.41",
"Cache-Control": "no-cache"
}
) as ws:
for msg in msgs:
self.date = formatdate() # Each message needs to have its send date
if not customspeak:
msg = self.mkssmlmsg(msg.decode('utf-8'), voice, pitch, rate, volume, customspeak=False)
else:
msg = self.mkssmlmsg(msg, customspeak=True)
message='X-Timestamp:'+self.date+'\r\nContent-Type:application/json; charset=utf-8\r\nPath:speech.config\r\n\r\n'
message+='{"context":{"synthesis":{"audio":{"metadataoptions":{"sentenceBoundaryEnabled":"'+sentenceBoundary+'","wordBoundaryEnabled":"'+wordBoundary+'"},"outputFormat":"' + codec + '"}}}}\r\n'
await ws.send(message)
await ws.send(msg)
download = False
async for recv in ws:
if type(recv) is str:
if 'turn.start' in recv:
download = True
elif 'turn.end' in recv:
download = False
break
elif 'audio.metadata' in recv:
#print("".join(recv.split('Path:audio.metadata\r\n\r\n')[1:]), file=sys.stderr)
metadata = json.loads("".join(recv.split('Path:audio.metadata\r\n\r\n')[1:]))
text = metadata['Metadata'][0]['Data']['text']['Text']
offset = metadata['Metadata'][0]['Data']['Offset']
yield [ offset, text, None ]
elif type(recv) is bytes:
if download:
yield [ None, None, b"".join(recv.split(b'Path:audio\r\n')[1:]) ]
await ws.close()
# Based on https://github.com/pndurette/gTTS/blob/6d9309f05b3ad26ca356654732f3b5b9c3bec538/gtts/utils.py#L13-L54
# Modified to measure based on bytes rather than number of characters
def _minimize(the_string, delim, max_size):
# Make sure we are measuring based on bytes
the_string = the_string.encode('utf-8') if type(the_string) is str else the_string
if the_string.startswith(delim):
the_string = the_string[len(delim):]
if len(the_string) > max_size:
try:
# Find the highest index of `delim` in `the_string[0:max_size]`
# i.e. `the_string` will be cut in half on `delim` index
idx = the_string.rindex(delim, 0, max_size)
except ValueError:
# `delim` not found in `the_string`, index becomes `max_size`
# i.e. `the_string` will be cut in half arbitrarily on `max_size`
idx = max_size
# Call itself again for `the_string[idx:]`
return [the_string[:idx]] + \
_minimize(the_string[idx:], delim, max_size)
else:
return [the_string]
async def _main():
parser = argparse.ArgumentParser(description="Microsoft Edge's Online TTS Reader")
group = parser.add_mutually_exclusive_group(required=True)
group.add_argument('-t', '--text', help='what TTS will say')
group.add_argument('-f', '--file', help='same as --text but read from file')
parser.add_argument(
"-L",
"--log-level",
default=logging.CRITICAL,
type=lambda x: getattr(logging, x),
help="configure the logging level (currently only DEBUG supported)"
)
parser.add_argument('-z', '--custom-ssml', help='treat text as ssml to send. For more info check https://bit.ly/3fIq13S', action='store_true')
parser.add_argument('-v', '--voice', help='voice for TTS. Default: Microsoft Server Speech Text to Speech Voice (en-US, AriaNeural)', default='Microsoft Server Speech Text to Speech Voice (en-US, AriaNeural)')
parser.add_argument('-c', '--codec', help="codec format. Default: audio-24khz-48kbitrate-mono-mp3. Another choice is webm-24khz-16bit-mono-opus. For more info check https://bit.ly/2T33h6S", default='audio-24khz-48kbitrate-mono-mp3')
group.add_argument('-l', '--list-voices', help="lists available voices. Edge's list is incomplete so check https://bit.ly/2SFq1d3", action='store_true')
parser.add_argument('-p', '--pitch', help="set TTS pitch. Default +0Hz, For more info check https://bit.ly/3eAE5Nx", default="+0Hz")
parser.add_argument('-r', '--rate', help="set TTS rate. Default +0%%. For more info check https://bit.ly/3eAE5Nx", default="+0%")
parser.add_argument('-V', '--volume', help="set TTS volume. Default +0%%. For more info check https://bit.ly/3eAE5Nx", default="+0%")
parser.add_argument('-s', '--enable-sentence-boundary', help="enable sentence boundary (not implemented but settable)", action='store_true')
parser.add_argument('-w', '--enable-word-boundary', help="enable word boundary (not implemented but settable)", action='store_true')
parser.add_argument('-O', '--overlapping', help="overlapping subtitles in seconds", default=5)
args = parser.parse_args()
logging.basicConfig(level=args.log_level)
logger = logging.getLogger("edgeTTS._main")
if (args.text or args.file) is not None:
if args.file is not None:
# we need to use sys.stdin.read() because some devices
# like Windows and Termux don't have a /dev/stdin.
if args.file == "/dev/stdin":
logger.debug("stdin detected, reading natively from stdin")
args.text = sys.stdin.read()
else:
logger.debug("reading from %s" % args.file)
with open(args.file, 'r') as file:
args.text = file.read()
tts = Communicate()
subs = SubMaker(args.overlapping)
async for i in tts.run(args.text, args.enable_sentence_boundary, args.enable_word_boundary, args.codec, args.voice, args.pitch, args.rate, args.volume, customspeak=args.custom_ssml):
if i[2] is not None:
sys.stdout.buffer.write(i[2])
elif (i[0] and i[1]) is not None:
subs.createSub(i[0], i[1])
if not subs.subsAndOffset == {}:
sys.stderr.write(subs.generateSubs())
elif args.list_voices:
seperator = False
for voice in list_voices():
if seperator: print()
for key in voice.keys():
logger.debug("Processing key %s" % key)
if key in ["SuggestedCodec", "FriendlyName", "Status"]:
logger.debug("Key %s skipped" % key)
continue
#print ("%s: %s" % ("Name" if key == "ShortName" else key, voice[key]))
print ("%s: %s" % (key, voice[key]))
seperator = True
def main():
asyncio.run(_main())
if __name__ == "__main__":
main()