tts.py

import argparse
import requests
import sys

TTS_SERVER_HOST = 'tts.mimi.fd.ai'
TTS_SERVER_PATH = '/v1/synthesis/nict'

def synthesize(token, input_lang, text, filename):
    headers = {
        'Authorization': f'Bearer {token}',
    }
    data = {
        'text': text,
        'lang': input_lang,
    }

    url = f'https://{TTS_SERVER_HOST}{TTS_SERVER_PATH}'
    resp = requests.post(url, headers=headers, data=data)
    resp.raise_for_status()
    with open(filename, 'wb') as fout:
        fout.write(resp.content)


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('access_token', help='access token file')
    parser.add_argument('input_lang', help='input language (e.g. ja)')
    parser.add_argument('input_text', help='input text')
    parser.add_argument('output_filename', help='output WAV filename')
    args = parser.parse_args()

    with open(args.access_token, 'r') as f:
        token = f.read().strip()

    print('start speech synthesis...', file=sys.stderr)
    try:
        synthesize(token, args.input_lang, args.input_text, args.output_filename)
    except Exception as e:
        print(f'ERROR: {type(e).__name__}: {e}', file=sys.stderr)
        sys.exit(1)
    print(f'saved wav file: {args.output_filename}', file=sys.stderr)