Alibaba Cloud Model Studio:SDK Python Sintesis Ucapan CosyVoice

# coding=utf-8

import dashscope
from dashscope.audio.tts_v2 import *
import os

# Kunci API berbeda antara wilayah Singapura dan Beijing. Dapatkan Kunci API Anda: https://www.alibabacloud.com/help/zh/model-studio/get-api-key
# Jika Anda belum menyetel variabel lingkungan, ganti baris berikut dengan Kunci API Model Studio Anda: dashscope.api_key = "sk-xxx"
dashscope.api_key = os.environ.get('DASHSCOPE_API_KEY')

# URL ini untuk wilayah Singapura. Untuk Beijing, ganti dengan: wss://dashscope.aliyuncs.com/api-ws/v1/inference
dashscope.base_websocket_api_url='wss://dashscope-intl.aliyuncs.com/api-ws/v1/inference'

# Model
model = "cosyvoice-v3-flash"
# Voice
voice = "longanyang"

# Buat instans SpeechSynthesizer dan kirim parameter permintaan seperti model dan voice di konstruktor
synthesizer = SpeechSynthesizer(model=model, voice=voice)
# Kirim teks untuk sintesis dan dapatkan audio biner
audio = synthesizer.call("How is the weather today?")
# Permintaan pertama membuat koneksi WebSocket, sehingga delay paket pertama mencakup waktu pembentukan koneksi
print('[Metric] requestId: {}, first-package delay: {} ms'.format(
    synthesizer.get_last_request_id(),
    synthesizer.get_first_package_delay()))

# Simpan audio ke file lokal
with open('output.mp3', 'wb') as f:
    f.write(audio)

# coding=utf-8

import os
import dashscope
from dashscope.audio.tts_v2 import *

from datetime import datetime

def get_timestamp():
    now = datetime.now()
    formatted_timestamp = now.strftime("[%Y-%m-%d %H:%M:%S.%f]")
    return formatted_timestamp

# Kunci API berbeda antara wilayah Singapura dan Beijing. Dapatkan Kunci API Anda: https://www.alibabacloud.com/help/zh/model-studio/get-api-key
# Jika Anda belum menyetel variabel lingkungan, ganti baris berikut dengan Kunci API Model Studio Anda: dashscope.api_key = "sk-xxx"
dashscope.api_key = os.environ.get('DASHSCOPE_API_KEY')

# URL ini untuk wilayah Singapura. Untuk Beijing, ganti dengan: wss://dashscope.aliyuncs.com/api-ws/v1/inference
dashscope.base_websocket_api_url='wss://dashscope-intl.aliyuncs.com/api-ws/v1/inference'

# Model
model = "cosyvoice-v3-flash"
# Voice
voice = "longanyang"


# Definisikan antarmuka callback
class Callback(ResultCallback):
    _player = None
    _stream = None

    def on_open(self):
        self.file = open("output.mp3", "wb")
        print("Koneksi terbentuk: " + get_timestamp())

    def on_complete(self):
        print("Sintesis ucapan selesai. Semua hasil diterima: " + get_timestamp())
        # Panggil get_first_package_delay hanya setelah on_complete dipicu
        # Permintaan pertama membuat koneksi WebSocket, sehingga delay paket pertama mencakup waktu pembentukan koneksi
        print('[Metric] requestId: {}, first-package delay: {} ms'.format(
            synthesizer.get_last_request_id(),
            synthesizer.get_first_package_delay()))

    def on_error(self, message: str):
        print(f"Kesalahan sintesis ucapan: {message}")

    def on_close(self):
        print("Koneksi ditutup: " + get_timestamp())
        self.file.close()

    def on_event(self, message):
        pass

    def on_data(self, data: bytes) -> None:
        print(get_timestamp() + " Panjang audio biner: " + str(len(data)))
        self.file.write(data)


callback = Callback()

# Buat instans SpeechSynthesizer dan kirim parameter permintaan seperti model dan voice di konstruktor
synthesizer = SpeechSynthesizer(
    model=model,
    voice=voice,
    callback=callback,
)

# Kirim teks untuk sintesis dan dapatkan audio biner secara real-time melalui metode on_data dari antarmuka callback
synthesizer.call("How is the weather today?")

# coding=utf-8
#
# Instal PyAudio:
# Di macOS, jalankan:
#   brew install portaudio
#   pip install pyaudio
# Di Debian/Ubuntu, jalankan:
#   sudo apt-get install python-pyaudio python3-pyaudio
#   atau
#   pip install pyaudio
# Di CentOS, jalankan:
#   sudo yum install -y portaudio portaudio-devel && pip install pyaudio
# Di Microsoft Windows, jalankan:
#   python -m pip install pyaudio

import os
import time
import pyaudio
import dashscope
from dashscope.api_entities.dashscope_response import SpeechSynthesisResponse
from dashscope.audio.tts_v2 import *

from datetime import datetime

def get_timestamp():
    now = datetime.now()
    formatted_timestamp = now.strftime("[%Y-%m-%d %H:%M:%S.%f]")
    return formatted_timestamp

# Kunci API berbeda antara wilayah Singapura dan Beijing. Dapatkan Kunci API Anda: https://www.alibabacloud.com/help/zh/model-studio/get-api-key
# Jika Anda belum menyetel variabel lingkungan, ganti baris berikut dengan Kunci API Model Studio Anda: dashscope.api_key = "sk-xxx"
dashscope.api_key = os.environ.get('DASHSCOPE_API_KEY')

# URL ini untuk wilayah Singapura. Untuk Beijing, ganti dengan: wss://dashscope.aliyuncs.com/api-ws/v1/inference
dashscope.base_websocket_api_url='wss://dashscope-intl.aliyuncs.com/api-ws/v1/inference'

# Model
model = "cosyvoice-v3-flash"
# Voice
voice = "longanyang"


# Definisikan antarmuka callback
class Callback(ResultCallback):
    _player = None
    _stream = None

    def on_open(self):
        print("Koneksi terbentuk: " + get_timestamp())
        self._player = pyaudio.PyAudio()
        self._stream = self._player.open(
            format=pyaudio.paInt16, channels=1, rate=22050, output=True
        )

    def on_complete(self):
        print("Sintesis ucapan selesai. Semua hasil diterima: " + get_timestamp())

    def on_error(self, message: str):
        print(f"Kesalahan sintesis ucapan: {message}")

    def on_close(self):
        print("Koneksi ditutup: " + get_timestamp())
        # Hentikan pemutar
        self._stream.stop_stream()
        self._stream.close()
        self._player.terminate()

    def on_event(self, message):
        pass

    def on_data(self, data: bytes) -> None:
        print(get_timestamp() + " Panjang audio biner: " + str(len(data)))
        self._stream.write(data)


callback = Callback()

test_text = [
    "Streaming text-to-speech SDK,",
    "converts input text",
    "into binary audio data.",
    "Compared with non-streaming TTS,",
    "streaming TTS offers better real-time performance.",
    "Users hear near-synchronous audio output",
    "as they type text,",
    "greatly improving interaction and reducing wait time.",
    "Ideal for large language model (LLM) scenarios",
    "where text is streamed for speech synthesis.",
]

# Buat instans SpeechSynthesizer dan kirim parameter permintaan seperti model dan voice di konstruktor
synthesizer = SpeechSynthesizer(
    model=model,
    voice=voice,
    format=AudioFormat.PCM_22050HZ_MONO_16BIT,  
    callback=callback,
)


# Streaming teks untuk sintesis. Dapatkan audio biner secara real-time melalui metode on_data dari antarmuka callback
for text in test_text:
    synthesizer.streaming_call(text)
    time.sleep(0.1)
# Akhiri sintesis ucapan streaming
synthesizer.streaming_complete()

# Permintaan pertama membuat koneksi WebSocket, sehingga delay paket pertama mencakup waktu pembentukan koneksi
print('[Metric] requestId: {}, first-package delay: {} ms'.format(
    synthesizer.get_last_request_id(),
    synthesizer.get_first_package_delay()))

class Callback(ResultCallback):
    def on_open(self) -> None:
        print('Terhubung')
    
    def on_data(self, data: bytes) -> None:
        # Tangani data audio biner

    def on_complete(self) -> None:
        print('Selesai')

    def on_error(self, message) -> None:
        print('Kesalahan:', message)

    def on_close(self) -> None:
        print('Koneksi ditutup')


callback = Callback()