1

有没有关于如何在 C 中使用 LAME API 的简单示例（带有解释）？我确实设法使用了以下代码（基于Is there any LAME c++ wraper\simplifier (working on Linux Mac and Win from pure code)?）：

FILE *pcm = fopen(input_file, "rb");
FILE *mp3 = fopen(output_file, "wb");
size_t nread;
int ret, nwrite;

// 1. Get lame version (OPTIONAL)
printf("Using LAME v%s\n", get_lame_version());

const int PCM_SIZE = 8192;
const int MP3_SIZE = 8192;

short pcm_buffer[PCM_SIZE * 2];
unsigned char mp3_buffer[MP3_SIZE];

// 2. Initializing
lame_t lame = lame_init();

// 3. Do some settings (OPTIONAL)
// lame_set_in_samplerate(lame, 48000);
lame_set_VBR(lame, vbr_default);
// lame_set_VBR_quality(lame, 2);

// 4. Initialize parameters
ret = lame_init_params(lame);
if (ret < 0) {
    printf("Error occurred during parameters initializing. Code = %d\n",
            ret);
    return 1;
}

do {
    // Read PCM_SIZE of array
    nread = fread(pcm_buffer, 2 * sizeof(short), PCM_SIZE, pcm);
    if (nread != 0) {
        // 5. Encode
        int nsamples = nread;
        short buffer_l[nsamples];
        short buffer_r[nsamples];

        printf("nread = %d\n", nread);
        printf("pcm_buffer.length = %d\n", sizeof(pcm_buffer)/sizeof(short));

        int j = 0;
        int i = 0;
        for (i = 0; i < nsamples; i++) {
            buffer_l[i] = pcm_buffer[j++];
            buffer_r[i] = pcm_buffer[j++];

        }

        nwrite = lame_encode_buffer(lame, buffer_l, buffer_r, nread,
                mp3_buffer, MP3_SIZE);
    } else {
        // 6. Flush and give some final frames
        nwrite = lame_encode_flush(lame, mp3_buffer, MP3_SIZE);
    }

    if (nwrite < 0) {
        printf("Error occurred during encoding. Code = %d\n", nwrite);
        return 1;
    }

    fwrite(mp3_buffer, nwrite, 1, mp3);
} while (nread != 0);

// 7. Write INFO tag (OPTIONAL)
// lame_mp3_tags_fid(lame, mp3);

// 8. Free internal data structures
lame_close(lame);

fclose(mp3);
fclose(pcm);

但问题是，

  1. 我不明白为什么要使用

    short pcm_buffer[PCM_SIZE * 2];
fread(pcm_buffer, 2 * sizeof(short), PCM_SIZE, pcm);

    代替

    short pcm_buffer[PCM_SIZE * 2];
fread(pcm_buffer, sizeof(short), PCM_SIZE * 2, pcm);

  2. 如果我必须使用第一种方式，如何在 Java 中使用 get 这样的 pcm_buffer？我想使用 JNI

  3. 这是获得 2 个缓冲区（左右）的正确方法吗？

谢谢

4

1 回答 1

0

1）我不明白为什么要使用

pcm_buffer为PCM_SIZE*2每个sizeof(short)字节大的元素（通常为 2 个字节）提供空间。即PCM_SIZE*2*sizeof(short)字节总数。

这就是为什么fread要读取PCM_SIZE的块2*sizeof(short)，它会填满缓冲区中的整个PCM_SIZE*2*sizeof(short)字节。

于 2012-07-14T16:44:34.247 回答