主页 / 问题 / 1576177
Accepted
QuL1ux
Asked:2024-04-15 00:43:47 +0800 CST2024-04-15 00:43:47 +0800 CST

收听discord.py的语音频道

  • 772

一般来说，情况是这样的，我希望机器人进入语音通道（如果有人在那里），并收听对话，将其记录在文件中以供进一步操作。进入语音通道一切都清楚了，但是如何听声音呢？我深入研究了文档并找到了它voice_client.source，但此代码未成功打印 None：

audio_data = ctx.voice_client.source
while True:
    print(audio_data)

据我了解，您需要创建一个单独的音频流作品，但我根本不知道如何做到这一点。谁能帮忙解答一下？

python-3.x

1 个回答

  1. Best Answer

    您可以使用图书馆PyCord。

    pip install py-cord

    在那里，该类VoiceClient有一个方法start_recording，允许您录制频道中用户的声音。

    有一次，我什至遇到了记录多个用户的问题，这与接收用户声音包的实现中的一些功能有关。在我报告问题后，对实现进行了一些更改，并将一个示例直接添加到库存储库中，介绍如何使用此方法并获取录音声音文件。

    有一个例子audio_recording.py- 用于记录一个用户

    还有一个记录多个用户的例子：audio_recording_merged.py

    第二个代码：

    import io

import pydub  # pip install pydub==0.25.1

import discord
from discord.sinks import MP3Sink

bot = discord.Bot()


@bot.event
async def on_ready():
    print(f"Logged in as {bot.user}")


async def finished_callback(sink: MP3Sink, channel: discord.TextChannel):
    mention_strs = []
    audio_segs: list[pydub.AudioSegment] = []
    files: list[discord.File] = []

    longest = pydub.AudioSegment.empty()

    for user_id, audio in sink.audio_data.items():
        mention_strs.append(f"<@{user_id}>")

        seg = pydub.AudioSegment.from_file(audio.file, format="mp3")

        # Determine the longest audio segment
        if len(seg) > len(longest):
            audio_segs.append(longest)
            longest = seg
        else:
            audio_segs.append(seg)

        audio.file.seek(0)
        files.append(discord.File(audio.file, filename=f"{user_id}.mp3"))

    for seg in audio_segs:
        longest = longest.overlay(seg)

    with io.BytesIO() as f:
        longest.export(f, format="mp3")
        await channel.send(
            f"Finished! Recorded audio for {', '.join(mention_strs)}.",
            files=files + [discord.File(f, filename="recording.mp3")],
        )


@bot.command()
async def join(ctx: discord.ApplicationContext):
    """Join the voice channel!"""
    voice = ctx.author.voice

    if not voice:
        return await ctx.respond("You're not in a vc right now")

    await voice.channel.connect()

    await ctx.respond("Joined!")


@bot.command()
async def start(ctx: discord.ApplicationContext):
    """Record the voice channel!"""
    voice = ctx.author.voice

    if not voice:
        return await ctx.respond("You're not in a vc right now")

    vc: discord.VoiceClient = ctx.voice_client

    if not vc:
        return await ctx.respond(
            "I'm not in a vc right now. Use `/join` to make me join!"
        )

    vc.start_recording(
        MP3Sink(),
        finished_callback,
        ctx.channel,
        sync_start=True,  # WARNING: This feature is very unstable and may break at any time.
    )

    await ctx.respond("The recording has started!")


@bot.command()
async def stop(ctx: discord.ApplicationContext):
    """Stop the recording"""
    vc: discord.VoiceClient = ctx.voice_client

    if not vc:
        return await ctx.respond("There's no recording going on right now")

    vc.stop_recording()

    await ctx.respond("The recording has stopped!")


@bot.command()
async def leave(ctx: discord.ApplicationContext):
    """Leave the voice channel!"""
    vc: discord.VoiceClient = ctx.voice_client

    if not vc:
        return await ctx.respond("I'm not in a vc right now")

    await vc.disconnect()

    await ctx.respond("Left!")


bot.run("TOKEN")

    研究它是如何工作的，阅读该方法的文档start_recording，我认为，您可以根据这个示例进行某种您自己的实现。

    • 1

相关问题